This is an R Markdown Notebook. When you execute code within the notebook, the results appear beneath the code.

Try executing this chunk by clicking the Run button within the chunk or by placing your cursor inside it and pressing Cmd+Shift+Enter.

## Installs
install.packages("stm")
trying URL 'https://cran.rstudio.com/bin/macosx/contrib/4.2/stm_1.3.6.tgz'
Content type 'application/x-gzip' length 3415835 bytes (3.3 MB)
==================================================
downloaded 3.3 MB

The downloaded binary packages are in
    /var/folders/82/tf8g1fnn6f19cftfxkss0w9c0000gn/T//RtmphRez9l/downloaded_packages
install.packages("quanteda")
trying URL 'https://cran.rstudio.com/bin/macosx/contrib/4.2/quanteda_3.3.0.tgz'
Content type 'application/x-gzip' length 3943481 bytes (3.8 MB)
==================================================
downloaded 3.8 MB

The downloaded binary packages are in
    /var/folders/82/tf8g1fnn6f19cftfxkss0w9c0000gn/T//RtmphRez9l/downloaded_packages
install.packages("spacyr")
trying URL 'https://cran.rstudio.com/bin/macosx/contrib/4.2/spacyr_1.2.1.tgz'
Content type 'application/x-gzip' length 268022 bytes (261 KB)
==================================================
downloaded 261 KB

The downloaded binary packages are in
    /var/folders/82/tf8g1fnn6f19cftfxkss0w9c0000gn/T//RtmphRez9l/downloaded_packages
install.packages("wordcloud")
trying URL 'https://cran.rstudio.com/bin/macosx/contrib/4.2/wordcloud_2.6.tgz'
Content type 'application/x-gzip' length 281705 bytes (275 KB)
==================================================
downloaded 275 KB

The downloaded binary packages are in
    /var/folders/82/tf8g1fnn6f19cftfxkss0w9c0000gn/T//RtmphRez9l/downloaded_packages
## Imports


library(jsonlite)
library(stm)
Registered S3 method overwritten by 'data.table':
  method           from
  print.data.table     
stm v1.3.6 successfully loaded. See ?stm for help. 
 Papers, resources, and other materials at structuraltopicmodel.com
library(quanteda)
Package version: 3.3.0
Unicode version: 14.0
ICU version: 70.1
Parallel computing: 8 of 8 threads used.
See https://quanteda.io for tutorials and examples.
library(spacyr)
library(stringr)
library(wordcloud)
Loading required package: RColorBrewer
library(ggplot2)
gender_func <- function(gender_label){
  if(gender_label == "woman"){
    return("Female")
  } else if(gender_label == "man") {
    return("Male")
  } else {
    return("other")
  }
}

w_vs <- function(eth_label){
  if(eth_label == 'white'){
    return("White")
  } else {
    return("Not White")
  }
}
a_vs <- function(eth_label){
  if(eth_label == 'asian'){
    return("Asian")
  } else {
    return("Not Asian")
  }
}

b_vs <- function(eth_label){
  if(eth_label == 'black'){
    return("African American")
  } else {
    return("Not African American")
  }
}
h_vs <- function(eth_label){
  if(eth_label == 'hispanic'){
    return("Hispanic")
  } else {
    return("Not Hispanic")
  }
}
# survey_meta <- read.csv("/Users/shinkamori/Documents/Research/p2345.csv")
# survey_meta <- read.csv("/Users/shinkamori/Documents/Research/p2_p4_p5_p6_p7.csv")
survey_meta <- read.csv("/Users/shinkamori/Documents/Research/p4_5_6_7.csv")
survey_meta$gender <- sapply(survey_meta$gender, gender_func)
# survey_meta$w_vs <- sapply(survey_meta$race, w_vs)
# survey_meta$a_vs <- sapply(survey_meta$race, a_vs)
# survey_meta$b_vs <- sapply(survey_meta$race, b_vs)
# survey_meta$h_vs <- sapply(survey_meta$race, h_vs)

# survey_meta <- survey_meta[, c("text", "gender", "race")]
summary(survey_meta)
       X           Unnamed..0        gender              race                tok           total.tok    
 Min.   :  0.0   Min.   :  0.00   Length:1496        Length:1496        Min.   : 58.74   Min.   :27.00  
 1st Qu.: 70.0   1st Qu.: 99.75   Class :character   Class :character   1st Qu.: 70.06   1st Qu.:28.00  
 Median :162.0   Median :199.50   Mode  :character   Mode  :character   Median : 82.94   Median :34.00  
 Mean   :175.2   Mean   :199.50                                         Mean   : 80.70   Mean   :32.17  
 3rd Qu.:275.0   3rd Qu.:299.25                                         3rd Qu.: 90.18   3rd Qu.:35.00  
 Max.   :399.0   Max.   :399.00                                         Max.   :102.74   Max.   :35.00  
                 NA's   :296                                                             NA's   :296    
     text              reason              w_vs               a_vs               b_vs          
 Length:1496        Length:1496        Length:1496        Length:1496        Length:1496       
 Class :character   Class :character   Class :character   Class :character   Class :character  
 Mode  :character   Mode  :character   Mode  :character   Mode  :character   Mode  :character  
                                                                                               
                                                                                               
                                                                                               
                                                                                               
     h_vs          
 Length:1496       
 Class :character  
 Mode  :character  
                   
                   
                   
                   
head(survey_meta)
sapply(lapply(survey_meta, unique), length)
  text gender   race 
  1485      2      4 
processed <- textProcessor(survey_meta$text, metadata=survey_meta[, c( "gender", "race")])
Building corpus... 
Converting to Lower Case... 
Removing punctuation... 
Removing stopwords... 
Removing numbers... 
Stemming... 
Creating Output... 
out <- prepDocuments(processed$documents, processed$vocab, processed$meta)
Removing 418 of 1310 terms (418 of 36615 tokens) due to frequency 
Your corpus now has 1440 documents, 892 terms and 36197 tokens.
sapply(lapply(processed$meta, unique), length)
gender   race 
     2      4 
temp_text <- survey_meta[-processed$docs.removed,]
length(temp_text)
[1] 3
model_depression_race <- stm::stm(
  documents = out_race$documents,
  K = 25,
  prevalence =~ gender + w_vs ,
  data=out_race$meta,
  vocab = out_race$vocab,
  max.em.its=75
)
Error in makeTopMatrix(prevalence, data) : Error creating model matrix.
                 This could be caused by many things including
                 explicit calls to a namespace within the formula.
                 Try a simpler formula.
model_depression_race_2 <- stm::stm(
  documents = out_race$documents,
  K = 25,
  prevalence =~ gender + race ,
  data=out_race$meta,
  vocab = out_race$vocab,
  max.em.its=75
)
Beginning Spectral Initialization 
     Calculating the gram matrix...
     Finding anchor words...
    .........................
     Recovering initialization...
    ........
Initialization complete.
..........................................................................................................
Completed E-Step (1 seconds). 
Completed M-Step. 
Completing Iteration 1 (approx. per word bound = -4.877) 
..........................................................................................................
Completed E-Step (0 seconds). 
Completed M-Step. 
Completing Iteration 2 (approx. per word bound = -4.707, relative change = 3.474e-02) 
..........................................................................................................
Completed E-Step (0 seconds). 
Completed M-Step. 
Completing Iteration 3 (approx. per word bound = -4.656, relative change = 1.082e-02) 
..........................................................................................................
Completed E-Step (0 seconds). 
Completed M-Step. 
Completing Iteration 4 (approx. per word bound = -4.635, relative change = 4.491e-03) 
..........................................................................................................
Completed E-Step (0 seconds). 
Completed M-Step. 
Completing Iteration 5 (approx. per word bound = -4.625, relative change = 2.175e-03) 
Topic 1: feel, like, enough, good, make 
 Topic 2: struggl, feel, find, depress, also 
 Topic 3: job, stress, work, life, sourc 
 Topic 4: feel, like, life, anyth, depress 
 Topic 5: realli, dont, like, know, think 
 Topic 6: feel, make, famili, also, financi 
 Topic 7: realli, feel, parent, want, alway 
 Topic 8: feel, enough, like, good, depress 
 Topic 9: thing, feel, depress, can, caus 
 Topic 10: stress, worri, job, work, sourc 
 Topic 11: feel, never, like, find, job 
 Topic 12: feel, constant, like, black, racism 
 Topic 13: just, right, feel, realli, now 
 Topic 14: feel, thing, make, constant, last 
 Topic 15: feel, realli, main, just, job 
 Topic 16: feel, like, well, enough, good 
 Topic 17: feel, get, stuck, dont, like 
 Topic 18: realli, feel, just, sure, late 
 Topic 19: feel, thing, like, make, last 
 Topic 20: feel, like, just, hard, tire 
 Topic 21: stress, realli, feel, sourc, life 
 Topic 22: stress, worri, life, feel, also 
 Topic 23: overwhelm, deal, person, cope, lot 
 Topic 24: alway, cant, feel, worri, just 
 Topic 25: feel, like, depress, may, thing 
..........................................................................................................
Completed E-Step (0 seconds). 
Completed M-Step. 
Completing Iteration 6 (approx. per word bound = -4.619, relative change = 1.412e-03) 
..........................................................................................................
Completed E-Step (0 seconds). 
Completed M-Step. 
Completing Iteration 7 (approx. per word bound = -4.614, relative change = 9.962e-04) 
..........................................................................................................
Completed E-Step (0 seconds). 
Completed M-Step. 
Completing Iteration 8 (approx. per word bound = -4.611, relative change = 7.578e-04) 
..........................................................................................................
Completed E-Step (0 seconds). 
Completed M-Step. 
Completing Iteration 9 (approx. per word bound = -4.608, relative change = 5.195e-04) 
..........................................................................................................
Completed E-Step (0 seconds). 
Completed M-Step. 
Completing Iteration 10 (approx. per word bound = -4.606, relative change = 4.067e-04) 
Topic 1: feel, like, enough, good, make 
 Topic 2: struggl, feel, find, depress, job 
 Topic 3: job, work, long, stress, hour 
 Topic 4: feel, like, life, live, anyth 
 Topic 5: realli, dont, think, know, man 
 Topic 6: make, worri, feel, money, financi 
 Topic 7: realli, want, parent, alway, famili 
 Topic 8: feel, enough, like, good, depress 
 Topic 9: thing, depress, feel, can, mani 
 Topic 10: stress, worri, job, work, sourc 
 Topic 11: never, feel, abl, find, job 
 Topic 12: constant, like, feel, black, racism 
 Topic 13: just, right, feel, realli, now 
 Topic 14: thing, feel, anoth, one, constant 
 Topic 15: main, feel, realli, late, just 
 Topic 16: feel, like, well, enough, good 
 Topic 17: get, feel, stuck, like, dont 
 Topic 18: realli, feel, just, sure, late 
 Topic 19: feel, thing, make, like, last 
 Topic 20: feel, like, tire, hard, just 
 Topic 21: stress, realli, take, health, sourc 
 Topic 22: worri, stress, futur, will, also 
 Topic 23: stress, overwhelm, can, deal, lot 
 Topic 24: alway, cant, keep, break, just 
 Topic 25: feel, like, depress, may, isol 
..........................................................................................................
Completed E-Step (0 seconds). 
Completed M-Step. 
Completing Iteration 11 (approx. per word bound = -4.605, relative change = 3.151e-04) 
..........................................................................................................
Completed E-Step (0 seconds). 
Completed M-Step. 
Completing Iteration 12 (approx. per word bound = -4.604, relative change = 2.402e-04) 
..........................................................................................................
Completed E-Step (0 seconds). 
Completed M-Step. 
Completing Iteration 13 (approx. per word bound = -4.603, relative change = 1.873e-04) 
..........................................................................................................
Completed E-Step (0 seconds). 
Completed M-Step. 
Completing Iteration 14 (approx. per word bound = -4.602, relative change = 1.591e-04) 
..........................................................................................................
Completed E-Step (0 seconds). 
Completed M-Step. 
Completing Iteration 15 (approx. per word bound = -4.602, relative change = 1.378e-04) 
Topic 1: feel, like, enough, good, make 
 Topic 2: struggl, feel, find, depress, job 
 Topic 3: job, long, work, hour, stress 
 Topic 4: feel, like, life, just, live 
 Topic 5: realli, dont, think, man, know 
 Topic 6: make, worri, money, feel, financi 
 Topic 7: parent, realli, want, alway, hispan 
 Topic 8: feel, enough, like, good, depress 
 Topic 9: thing, depress, feel, can, mani 
 Topic 10: stress, work, job, sourc, worri 
 Topic 11: never, feel, abl, job, find 
 Topic 12: constant, like, black, racism, feel 
 Topic 13: just, right, feel, realli, now 
 Topic 14: thing, anoth, feel, one, relationship 
 Topic 15: main, feel, realli, work, late 
 Topic 16: feel, like, well, enough, good 
 Topic 17: get, feel, stuck, like, dont 
 Topic 18: realli, sure, feel, just, late 
 Topic 19: feel, thing, make, like, last 
 Topic 20: feel, like, tire, hard, just 
 Topic 21: realli, stress, take, health, mental 
 Topic 22: worri, futur, will, also, stress 
 Topic 23: stress, lot, can, overwhelm, deal 
 Topic 24: keep, alway, cant, break, just 
 Topic 25: feel, like, depress, isol, may 
..........................................................................................................
Completed E-Step (0 seconds). 
Completed M-Step. 
Completing Iteration 16 (approx. per word bound = -4.601, relative change = 1.303e-04) 
..........................................................................................................
Completed E-Step (0 seconds). 
Completed M-Step. 
Completing Iteration 17 (approx. per word bound = -4.600, relative change = 1.166e-04) 
..........................................................................................................
Completed E-Step (0 seconds). 
Completed M-Step. 
Completing Iteration 18 (approx. per word bound = -4.600, relative change = 1.082e-04) 
..........................................................................................................
Completed E-Step (0 seconds). 
Completed M-Step. 
Completing Iteration 19 (approx. per word bound = -4.600, relative change = 9.000e-05) 
..........................................................................................................
Completed E-Step (0 seconds). 
Completed M-Step. 
Completing Iteration 20 (approx. per word bound = -4.599, relative change = 7.671e-05) 
Topic 1: feel, like, perfect, make, good 
 Topic 2: struggl, feel, find, depress, job 
 Topic 3: job, long, hour, work, stress 
 Topic 4: feel, like, life, just, live 
 Topic 5: realli, dont, think, man, asian 
 Topic 6: worri, make, money, financi, end 
 Topic 7: hispan, parent, alway, famili, realli 
 Topic 8: feel, enough, like, good, depress 
 Topic 9: thing, depress, feel, can, mani 
 Topic 10: stress, work, sourc, job, moment 
 Topic 11: never, abl, job, feel, find 
 Topic 12: constant, racism, like, black, feel 
 Topic 13: just, right, feel, realli, now 
 Topic 14: thing, anoth, feel, one, relationship 
 Topic 15: main, stress, realli, feel, work 
 Topic 16: feel, like, well, enough, good 
 Topic 17: get, feel, stuck, like, dont 
 Topic 18: realli, sure, feel, just, late 
 Topic 19: feel, thing, make, like, last 
 Topic 20: feel, like, tire, hard, just 
 Topic 21: realli, stress, take, health, mental 
 Topic 22: worri, futur, will, also, sure 
 Topic 23: stress, lot, can, overwhelm, deal 
 Topic 24: keep, cant, alway, break, just 
 Topic 25: feel, isol, like, depress, may 
..........................................................................................................
Completed E-Step (0 seconds). 
Completed M-Step. 
Completing Iteration 21 (approx. per word bound = -4.599, relative change = 6.319e-05) 
..........................................................................................................
Completed E-Step (0 seconds). 
Completed M-Step. 
Completing Iteration 22 (approx. per word bound = -4.599, relative change = 5.931e-05) 
..........................................................................................................
Completed E-Step (0 seconds). 
Completed M-Step. 
Completing Iteration 23 (approx. per word bound = -4.598, relative change = 4.147e-05) 
..........................................................................................................
Completed E-Step (0 seconds). 
Completed M-Step. 
Completing Iteration 24 (approx. per word bound = -4.598, relative change = 4.456e-05) 
..........................................................................................................
Completed E-Step (0 seconds). 
Completed M-Step. 
Completing Iteration 25 (approx. per word bound = -4.598, relative change = 4.105e-05) 
Topic 1: feel, like, perfect, alway, make 
 Topic 2: struggl, feel, depress, find, job 
 Topic 3: job, long, hour, work, life 
 Topic 4: feel, like, life, just, live 
 Topic 5: realli, dont, think, man, asian 
 Topic 6: worri, make, money, financi, end 
 Topic 7: hispan, famili, parent, alway, realli 
 Topic 8: enough, feel, like, good, depress 
 Topic 9: thing, depress, feel, can, mani 
 Topic 10: stress, sourc, work, moment, job 
 Topic 11: never, abl, find, job, feel 
 Topic 12: constant, racism, like, black, feel 
 Topic 13: just, right, feel, realli, now 
 Topic 14: thing, anoth, one, feel, relationship 
 Topic 15: stress, main, realli, life, feel 
 Topic 16: feel, like, well, enough, pressur 
 Topic 17: get, feel, stuck, like, dont 
 Topic 18: realli, sure, feel, just, what 
 Topic 19: feel, thing, make, last, like 
 Topic 20: feel, like, tire, just, hard 
 Topic 21: realli, take, health, stress, mental 
 Topic 22: worri, futur, will, also, sure 
 Topic 23: stress, lot, can, overwhelm, deal 
 Topic 24: keep, cant, break, alway, seem 
 Topic 25: feel, isol, depress, may, like 
..........................................................................................................
Completed E-Step (0 seconds). 
Completed M-Step. 
Completing Iteration 26 (approx. per word bound = -4.598, relative change = 4.110e-05) 
..........................................................................................................
Completed E-Step (0 seconds). 
Completed M-Step. 
Completing Iteration 27 (approx. per word bound = -4.598, relative change = 3.227e-05) 
..........................................................................................................
Completed E-Step (0 seconds). 
Completed M-Step. 
Completing Iteration 28 (approx. per word bound = -4.598, relative change = 2.559e-05) 
..........................................................................................................
Completed E-Step (0 seconds). 
Completed M-Step. 
Completing Iteration 29 (approx. per word bound = -4.597, relative change = 2.605e-05) 
..........................................................................................................
Completed E-Step (0 seconds). 
Completed M-Step. 
Completing Iteration 30 (approx. per word bound = -4.597, relative change = 1.845e-05) 
Topic 1: feel, like, alway, perfect, time 
 Topic 2: struggl, feel, depress, job, find 
 Topic 3: job, long, hour, work, life 
 Topic 4: feel, like, life, just, live 
 Topic 5: realli, dont, think, man, asian 
 Topic 6: worri, make, money, financi, end 
 Topic 7: hispan, famili, parent, alway, realli 
 Topic 8: enough, feel, like, good, depress 
 Topic 9: thing, depress, feel, can, problem 
 Topic 10: stress, sourc, work, moment, job 
 Topic 11: never, abl, find, job, feel 
 Topic 12: constant, racism, like, black, feel 
 Topic 13: just, right, feel, realli, now 
 Topic 14: thing, anoth, one, relationship, feel 
 Topic 15: stress, main, realli, life, job 
 Topic 16: feel, like, well, enough, one 
 Topic 17: get, feel, like, stuck, dont 
 Topic 18: realli, sure, feel, what, just 
 Topic 19: feel, thing, make, last, like 
 Topic 20: feel, like, tire, just, hard 
 Topic 21: realli, take, health, mental, toll 
 Topic 22: worri, futur, will, also, life 
 Topic 23: stress, lot, can, deal, overwhelm 
 Topic 24: keep, cant, break, seem, everyth 
 Topic 25: feel, isol, may, depress, overwhelm 
..........................................................................................................
Completed E-Step (0 seconds). 
Completed M-Step. 
Completing Iteration 31 (approx. per word bound = -4.597, relative change = 1.369e-05) 
..........................................................................................................
Completed E-Step (0 seconds). 
Completed M-Step. 
Completing Iteration 32 (approx. per word bound = -4.597, relative change = 1.491e-05) 
..........................................................................................................
Completed E-Step (0 seconds). 
Completed M-Step. 
Model Converged 
model_depression <- stm::stm(
  documents = out$documents,
  K = 25,
  prevalence =~ gender + race,
  data=out$meta,
  vocab = out$vocab,
  max.em.its=75
)
Beginning Spectral Initialization 
     Calculating the gram matrix...
     Finding anchor words...
    .........................
     Recovering initialization...
    ........
Initialization complete.
..........................................................................................................
Completed E-Step (1 seconds). 
Completed M-Step. 
Completing Iteration 1 (approx. per word bound = -4.877) 
..........................................................................................................
Completed E-Step (0 seconds). 
Completed M-Step. 
Completing Iteration 2 (approx. per word bound = -4.707, relative change = 3.474e-02) 
..........................................................................................................
Completed E-Step (0 seconds). 
Completed M-Step. 
Completing Iteration 3 (approx. per word bound = -4.656, relative change = 1.082e-02) 
..........................................................................................................
Completed E-Step (0 seconds). 
Completed M-Step. 
Completing Iteration 4 (approx. per word bound = -4.635, relative change = 4.491e-03) 
..........................................................................................................
Completed E-Step (0 seconds). 
Completed M-Step. 
Completing Iteration 5 (approx. per word bound = -4.625, relative change = 2.175e-03) 
Topic 1: feel, like, enough, good, make 
 Topic 2: struggl, feel, find, depress, also 
 Topic 3: job, stress, work, life, sourc 
 Topic 4: feel, like, life, anyth, depress 
 Topic 5: realli, dont, like, know, think 
 Topic 6: feel, make, famili, also, financi 
 Topic 7: realli, feel, parent, want, alway 
 Topic 8: feel, enough, like, good, depress 
 Topic 9: thing, feel, depress, can, caus 
 Topic 10: stress, worri, job, work, sourc 
 Topic 11: feel, never, like, find, job 
 Topic 12: feel, constant, like, black, racism 
 Topic 13: just, right, feel, realli, now 
 Topic 14: feel, thing, make, constant, last 
 Topic 15: feel, realli, main, just, job 
 Topic 16: feel, like, well, enough, good 
 Topic 17: feel, get, stuck, dont, like 
 Topic 18: realli, feel, just, sure, late 
 Topic 19: feel, thing, like, make, last 
 Topic 20: feel, like, just, hard, tire 
 Topic 21: stress, realli, feel, sourc, life 
 Topic 22: stress, worri, life, feel, also 
 Topic 23: overwhelm, deal, person, cope, lot 
 Topic 24: alway, cant, feel, worri, just 
 Topic 25: feel, like, depress, may, thing 
..........................................................................................................
Completed E-Step (0 seconds). 
Completed M-Step. 
Completing Iteration 6 (approx. per word bound = -4.619, relative change = 1.412e-03) 
..........................................................................................................
Completed E-Step (0 seconds). 
Completed M-Step. 
Completing Iteration 7 (approx. per word bound = -4.614, relative change = 9.962e-04) 
..........................................................................................................
Completed E-Step (0 seconds). 
Completed M-Step. 
Completing Iteration 8 (approx. per word bound = -4.611, relative change = 7.578e-04) 
..........................................................................................................
Completed E-Step (0 seconds). 
Completed M-Step. 
Completing Iteration 9 (approx. per word bound = -4.608, relative change = 5.195e-04) 
..........................................................................................................
Completed E-Step (0 seconds). 
Completed M-Step. 
Completing Iteration 10 (approx. per word bound = -4.606, relative change = 4.067e-04) 
Topic 1: feel, like, enough, good, make 
 Topic 2: struggl, feel, find, depress, job 
 Topic 3: job, work, long, stress, hour 
 Topic 4: feel, like, life, live, anyth 
 Topic 5: realli, dont, think, know, man 
 Topic 6: make, worri, feel, money, financi 
 Topic 7: realli, want, parent, alway, famili 
 Topic 8: feel, enough, like, good, depress 
 Topic 9: thing, depress, feel, can, mani 
 Topic 10: stress, worri, job, work, sourc 
 Topic 11: never, feel, abl, find, job 
 Topic 12: constant, like, feel, black, racism 
 Topic 13: just, right, feel, realli, now 
 Topic 14: thing, feel, anoth, one, constant 
 Topic 15: main, feel, realli, late, just 
 Topic 16: feel, like, well, enough, good 
 Topic 17: get, feel, stuck, like, dont 
 Topic 18: realli, feel, just, sure, late 
 Topic 19: feel, thing, make, like, last 
 Topic 20: feel, like, tire, hard, just 
 Topic 21: stress, realli, take, health, sourc 
 Topic 22: worri, stress, futur, will, also 
 Topic 23: stress, overwhelm, can, deal, lot 
 Topic 24: alway, cant, keep, break, just 
 Topic 25: feel, like, depress, may, isol 
..........................................................................................................
Completed E-Step (0 seconds). 
Completed M-Step. 
Completing Iteration 11 (approx. per word bound = -4.605, relative change = 3.151e-04) 
..........................................................................................................
Completed E-Step (0 seconds). 
Completed M-Step. 
Completing Iteration 12 (approx. per word bound = -4.604, relative change = 2.402e-04) 
..........................................................................................................
Completed E-Step (0 seconds). 
Completed M-Step. 
Completing Iteration 13 (approx. per word bound = -4.603, relative change = 1.873e-04) 
..........................................................................................................
Completed E-Step (0 seconds). 
Completed M-Step. 
Completing Iteration 14 (approx. per word bound = -4.602, relative change = 1.591e-04) 
..........................................................................................................
Completed E-Step (0 seconds). 
Completed M-Step. 
Completing Iteration 15 (approx. per word bound = -4.602, relative change = 1.378e-04) 
Topic 1: feel, like, enough, good, make 
 Topic 2: struggl, feel, find, depress, job 
 Topic 3: job, long, work, hour, stress 
 Topic 4: feel, like, life, just, live 
 Topic 5: realli, dont, think, man, know 
 Topic 6: make, worri, money, feel, financi 
 Topic 7: parent, realli, want, alway, hispan 
 Topic 8: feel, enough, like, good, depress 
 Topic 9: thing, depress, feel, can, mani 
 Topic 10: stress, work, job, sourc, worri 
 Topic 11: never, feel, abl, job, find 
 Topic 12: constant, like, black, racism, feel 
 Topic 13: just, right, feel, realli, now 
 Topic 14: thing, anoth, feel, one, relationship 
 Topic 15: main, feel, realli, work, late 
 Topic 16: feel, like, well, enough, good 
 Topic 17: get, feel, stuck, like, dont 
 Topic 18: realli, sure, feel, just, late 
 Topic 19: feel, thing, make, like, last 
 Topic 20: feel, like, tire, hard, just 
 Topic 21: realli, stress, take, health, mental 
 Topic 22: worri, futur, will, also, stress 
 Topic 23: stress, lot, can, overwhelm, deal 
 Topic 24: keep, alway, cant, break, just 
 Topic 25: feel, like, depress, isol, may 
..........................................................................................................
Completed E-Step (0 seconds). 
Completed M-Step. 
Completing Iteration 16 (approx. per word bound = -4.601, relative change = 1.303e-04) 
..........................................................................................................
Completed E-Step (0 seconds). 
Completed M-Step. 
Completing Iteration 17 (approx. per word bound = -4.600, relative change = 1.166e-04) 
..........................................................................................................
Completed E-Step (0 seconds). 
Completed M-Step. 
Completing Iteration 18 (approx. per word bound = -4.600, relative change = 1.082e-04) 
..........................................................................................................
Completed E-Step (0 seconds). 
Completed M-Step. 
Completing Iteration 19 (approx. per word bound = -4.600, relative change = 9.000e-05) 
..........................................................................................................
Completed E-Step (0 seconds). 
Completed M-Step. 
Completing Iteration 20 (approx. per word bound = -4.599, relative change = 7.671e-05) 
Topic 1: feel, like, perfect, make, good 
 Topic 2: struggl, feel, find, depress, job 
 Topic 3: job, long, hour, work, stress 
 Topic 4: feel, like, life, just, live 
 Topic 5: realli, dont, think, man, asian 
 Topic 6: worri, make, money, financi, end 
 Topic 7: hispan, parent, alway, famili, realli 
 Topic 8: feel, enough, like, good, depress 
 Topic 9: thing, depress, feel, can, mani 
 Topic 10: stress, work, sourc, job, moment 
 Topic 11: never, abl, job, feel, find 
 Topic 12: constant, racism, like, black, feel 
 Topic 13: just, right, feel, realli, now 
 Topic 14: thing, anoth, feel, one, relationship 
 Topic 15: main, stress, realli, feel, work 
 Topic 16: feel, like, well, enough, good 
 Topic 17: get, feel, stuck, like, dont 
 Topic 18: realli, sure, feel, just, late 
 Topic 19: feel, thing, make, like, last 
 Topic 20: feel, like, tire, hard, just 
 Topic 21: realli, stress, take, health, mental 
 Topic 22: worri, futur, will, also, sure 
 Topic 23: stress, lot, can, overwhelm, deal 
 Topic 24: keep, cant, alway, break, just 
 Topic 25: feel, isol, like, depress, may 
..........................................................................................................
Completed E-Step (0 seconds). 
Completed M-Step. 
Completing Iteration 21 (approx. per word bound = -4.599, relative change = 6.319e-05) 
..........................................................................................................
Completed E-Step (0 seconds). 
Completed M-Step. 
Completing Iteration 22 (approx. per word bound = -4.599, relative change = 5.931e-05) 
..........................................................................................................
Completed E-Step (0 seconds). 
Completed M-Step. 
Completing Iteration 23 (approx. per word bound = -4.598, relative change = 4.147e-05) 
..........................................................................................................
Completed E-Step (0 seconds). 
Completed M-Step. 
Completing Iteration 24 (approx. per word bound = -4.598, relative change = 4.456e-05) 
..........................................................................................................
Completed E-Step (0 seconds). 
Completed M-Step. 
Completing Iteration 25 (approx. per word bound = -4.598, relative change = 4.105e-05) 
Topic 1: feel, like, perfect, alway, make 
 Topic 2: struggl, feel, depress, find, job 
 Topic 3: job, long, hour, work, life 
 Topic 4: feel, like, life, just, live 
 Topic 5: realli, dont, think, man, asian 
 Topic 6: worri, make, money, financi, end 
 Topic 7: hispan, famili, parent, alway, realli 
 Topic 8: enough, feel, like, good, depress 
 Topic 9: thing, depress, feel, can, mani 
 Topic 10: stress, sourc, work, moment, job 
 Topic 11: never, abl, find, job, feel 
 Topic 12: constant, racism, like, black, feel 
 Topic 13: just, right, feel, realli, now 
 Topic 14: thing, anoth, one, feel, relationship 
 Topic 15: stress, main, realli, life, feel 
 Topic 16: feel, like, well, enough, pressur 
 Topic 17: get, feel, stuck, like, dont 
 Topic 18: realli, sure, feel, just, what 
 Topic 19: feel, thing, make, last, like 
 Topic 20: feel, like, tire, just, hard 
 Topic 21: realli, take, health, stress, mental 
 Topic 22: worri, futur, will, also, sure 
 Topic 23: stress, lot, can, overwhelm, deal 
 Topic 24: keep, cant, break, alway, seem 
 Topic 25: feel, isol, depress, may, like 
..........................................................................................................
Completed E-Step (0 seconds). 
Completed M-Step. 
Completing Iteration 26 (approx. per word bound = -4.598, relative change = 4.110e-05) 
..........................................................................................................
Completed E-Step (0 seconds). 
Completed M-Step. 
Completing Iteration 27 (approx. per word bound = -4.598, relative change = 3.227e-05) 
..........................................................................................................
Completed E-Step (0 seconds). 
Completed M-Step. 
Completing Iteration 28 (approx. per word bound = -4.598, relative change = 2.559e-05) 
..........................................................................................................
Completed E-Step (0 seconds). 
Completed M-Step. 
Completing Iteration 29 (approx. per word bound = -4.597, relative change = 2.605e-05) 
..........................................................................................................
Completed E-Step (0 seconds). 
Completed M-Step. 
Completing Iteration 30 (approx. per word bound = -4.597, relative change = 1.845e-05) 
Topic 1: feel, like, alway, perfect, time 
 Topic 2: struggl, feel, depress, job, find 
 Topic 3: job, long, hour, work, life 
 Topic 4: feel, like, life, just, live 
 Topic 5: realli, dont, think, man, asian 
 Topic 6: worri, make, money, financi, end 
 Topic 7: hispan, famili, parent, alway, realli 
 Topic 8: enough, feel, like, good, depress 
 Topic 9: thing, depress, feel, can, problem 
 Topic 10: stress, sourc, work, moment, job 
 Topic 11: never, abl, find, job, feel 
 Topic 12: constant, racism, like, black, feel 
 Topic 13: just, right, feel, realli, now 
 Topic 14: thing, anoth, one, relationship, feel 
 Topic 15: stress, main, realli, life, job 
 Topic 16: feel, like, well, enough, one 
 Topic 17: get, feel, like, stuck, dont 
 Topic 18: realli, sure, feel, what, just 
 Topic 19: feel, thing, make, last, like 
 Topic 20: feel, like, tire, just, hard 
 Topic 21: realli, take, health, mental, toll 
 Topic 22: worri, futur, will, also, life 
 Topic 23: stress, lot, can, deal, overwhelm 
 Topic 24: keep, cant, break, seem, everyth 
 Topic 25: feel, isol, may, depress, overwhelm 
..........................................................................................................
Completed E-Step (0 seconds). 
Completed M-Step. 
Completing Iteration 31 (approx. per word bound = -4.597, relative change = 1.369e-05) 
..........................................................................................................
Completed E-Step (0 seconds). 
Completed M-Step. 
Completing Iteration 32 (approx. per word bound = -4.597, relative change = 1.491e-05) 
..........................................................................................................
Completed E-Step (0 seconds). 
Completed M-Step. 
Model Converged 
labelTopics(model_depression_race_2, n = 30)
Topic 1 Top Words:
     Highest Prob: feel, like, alway, perfect, time, make, never, good, mistak, enough, can, depress, cant, everyth, relax, allow, peopl, pressur, ever, human, measur, friend, let, love, respect, compar, alon, anyon, reach, famili 
     FREX: perfect, mistak, time, allow, alway, relax, human, never, respect, let, good, measur, make, like, ever, everyth, progress, feel, can, cant, enough, full, love, reach, peopl, daughter, less, worthi, mind, compar 
     Lift: allow, perfect, human, bodi, daughter, mind, respect, mistak, relax, notic, contact, capabl, attent, let, truli, cost, possibl, appear, sister, ever, time, progress, less, alway, exercis, never, full, measur, student, outsid 
     Score: allow, perfect, mistak, alway, never, relax, human, enough, cant, good, like, can, make, time, feel, everyth, measur, ever, depress, let, respect, pressur, compar, full, worthi, truli, friend, peopl, reach, progress 
Topic 2 Top Words:
     Highest Prob: struggl, feel, depress, job, find, also, famili, relationship, friend, health, enjoy, mental, like, provid, miss, partner, sever, home, behind, back, fail, doesnt, first, place, support, alon, adjust, move, rent, love 
     FREX: struggl, find, miss, enjoy, partner, back, friend, home, relationship, rent, homesick, adjust, famili, also, diagnos, fail, class, doesnt, sever, provid, foremost, job, depress, evict, health, mental, behind, move, profession, place 
     Lift: class, homesick, diagnos, rent, adjust, evict, foremost, symptom, part-tim, romant, profession, medic, back, struggl, partner, third, barrier, fun, find, asia, sever, miss, doesnt, home, enjoy, relationship, student, appear, move, fail 
     Score: struggl, class, find, job, relationship, depress, health, mental, famili, also, partner, friend, enjoy, homesick, provid, diagnos, miss, home, sever, fail, rent, first, back, adjust, symptom, evict, behind, foremost, move, profession 
Topic 3 Top Words:
     Highest Prob: job, long, hour, work, life, stress, sourc, also, biggest, worri, moment, lot, famili, paid, well, dont, get, respons, time, good, much, abl, constant, sometim, lose, futur, sure, say, support, fire 
     FREX: hour, long, paid, biggest, job, moment, sourc, work, respons, lose, also, sole, spend, breadwinn, life, laid, worri, microscop, fire, factori, immigr, much, afford, stress, well, injur, famili, lot, home, status 
     Lift: injur, visa, construct, worker, paid, sole, weekend, microscop, hour, breadwinn, english, long, respons, lose, spend, speak, factori, cost, afford, laid, sibl, immigr, communic, incom, carri, comfort, decis, goe, sight, healthi 
     Score: hour, long, biggest, sourc, paid, moment, job, stress, respons, worri, visa, work, also, well, life, lot, famili, immigr, breadwinn, lose, fire, status, sole, much, say, factori, microscop, injur, spend, abl 
Topic 4 Top Words:
     Highest Prob: feel, like, life, just, live, anyth, depress, motion, realli, potenti, purpos, dont, exist, want, fact, day, accomplish, without, peopl, wast, reach, look, forward, hobbi, real, passion, full, goal, differ, make 
     FREX: purpos, motion, accomplish, potenti, live, exist, without, anyth, forward, passion, life, real, wast, float, reach, hobbi, fact, meaning, drift, like, full, just, day, goal, feel, ambit, mean, fullest, connect, look 
     Lift: ambit, purpos, drift, meaning, passion, float, accomplish, forward, motion, without, fullest, real, space, exist, hobbi, potenti, wast, goal, full, reach, mean, live, kind, connect, noth, hes, necessarili, puppet, away, interest 
     Score: anyth, motion, ambit, purpos, potenti, live, life, just, exist, accomplish, wast, like, real, depress, feel, forward, meaning, float, reach, without, day, passion, full, fact, hobbi, drift, direct, look, anywher, goal 
Topic 5 Top Words:
     Highest Prob: realli, dont, think, man, asian, depress, feel, woman, know, like, white, way, belong, late, turn, help, alon, place, just, need, societi, hispan, world, cultur, there, anywher, fit, old, better, year 
     FREX: asian, man, belong, think, woman, turn, white, fit, old, cultur, societi, help, place, dont, sorri, dark, way, hello, born, alon, failur, know, anywher, whos, there, sound, need, realli, either, american 
     Lift: middl, privileg, whos, sound, american, asian, belong, old, fit, born, sexism, man, cultur, hello, sorri, group, either, dark, failur, bad, insid, think, societi, express, histori, woman, white, turn, place, mexico 
     Score: asian, man, privileg, think, belong, white, turn, know, dont, woman, realli, old, fit, cultur, societi, anywher, depress, dark, sorri, born, way, year, whos, help, hispan, middl, late, hello, failur, world 
Topic 6 Top Words:
     Highest Prob: worri, make, money, financi, end, meet, feel, pay, bill, situat, constant, famili, also, one, work, provid, abl, mani, futur, current, support, save, two, bare, insecur, polit, reli, communiti, woman, whether 
     FREX: end, money, pay, financi, worri, bill, meet, situat, make, reli, insecur, provid, bare, save, threaten, polit, two, mother, date, current, attack, support, pandem, famili, member, mani, constant, communiti, futur, one 
     Lift: pandem, date, reli, end, threaten, insecur, pay, bill, situat, money, financi, bare, full-tim, mother, burnt, save, latina, abil, cover, scrape, expens, yet, attack, member, ident, provid, recogn, either, two, meet 
     Score: money, financi, end, worri, pay, bill, meet, pandem, situat, make, provid, reli, bare, mani, insecur, mother, two, save, polit, current, constant, threaten, member, futur, abl, attack, famili, date, children, full-tim 
Topic 7 Top Words:
     Highest Prob: hispan, famili, alway, parent, realli, want, hard, live, everyon, get, come, hey, ive, share, now, work, two, expect, what, came, year, right, ago, kid, america, tradit, countri, just, grew, disappoint 
     FREX: parent, hispan, came, come, tradit, ago, grade, grew, kid, two, share, strict, want, countri, year, hey, famili, child, marri, everyon, deport, proud, immigr, america, household, felt, still, disappoint, older, live 
     Lift: adult, strict, undocu, came, didnt, household, shadow, stabl, tradit, child, brother, beat, older, grade, parent, grew, caught, ago, hispan, proud, marri, larg, educ, fulli, kid, sister, come, colleg, languag, deport 
     Score: hispan, undocu, parent, came, grew, tradit, hey, ago, share, strict, grade, year, two, alway, famili, deport, immigr, come, everyon, felt, want, countri, what, household, proud, america, status, child, ive, live 
Topic 8 Top Words:
     Highest Prob: enough, feel, like, good, depress, anyth, never, ill, success, smart, pretti, worth, happi, compar, measur, worthi, other, will, burden, everyon, women, deserv, love, talent, one, just, help, cut, thought, see 
     FREX: enough, smart, success, good, pretti, worthi, worth, anyth, talent, burden, measur, ill, women, deserv, skinni, like, compar, happi, feel, depress, other, handsom, never, thought, thin, beauti, screw, shouldnt, cut, nobodi 
     Lift: thin, smart, talent, skinni, handsom, burden, worthi, success, nobodi, worth, pretti, good, deserv, enough, women, thought, beauti, worthless, measur, screw, inferior, compet, ill, compar, other, awhil, shouldnt, attract, cut, anyth 
     Score: enough, good, smart, anyth, thin, success, like, worth, ill, depress, worthi, feel, measur, compar, burden, never, talent, skinni, pretti, women, deserv, screw, other, shouldnt, worthless, happi, will, thought, beauti, handsom 
Topic 9 Top Words:
     Highest Prob: thing, depress, feel, can, problem, mani, caus, combin, person, might, contribut, troubl, number, also, relationship, discourag, way, hopeless, often, mayb, life, reason, think, talk, factor, addit, therapist, due, may, help 
     FREX: problem, number, combin, mayb, contribut, factor, common, mani, includ, troubl, discourag, therapist, due, might, addit, thing, hopeless, can, isnt, pinpoint, reason, inadequaci, caus, whatev, trigger, often, person, difficulti, chronic, women 
     Lift: chronic, address, brain, common, doctor, lay, exact, extrem, number, includ, activ, factor, dishearten, mayb, problem, inadequaci, due, therapist, most, contribut, combin, isnt, pinpoint, trigger, mood, addit, usual, mani, condit, whatev 
     Score: problem, combin, address, thing, mani, contribut, common, depress, can, caus, mayb, number, therapist, troubl, might, includ, due, may, person, factor, whatev, chronic, addit, reason, pinpoint, relationship, often, lead, discourag, difficulti 
Topic 10 Top Words:
     Highest Prob: stress, sourc, work, moment, job, like, life, biggest, worri, feel, constant, pressur, also, environ, meet, often, enough, perform, whether, high-pressur, lot, alway, futur, abl, well, deadlin, expect, mistak, sometim, make 
     FREX: environ, moment, biggest, high-pressur, perform, sourc, often, pressur, whether, deadlin, meet, work, fast-pac, stress, worri, job, bosss, demand, mistak, retir, constant, team, fire, life, expect, demot, corpor, also, say, probabl 
     Lift: graduat, environ, fast-pac, high-pressur, scrutini, debilit, deadlin, perform, demot, team, often, bosss, retir, whether, second-guess, level, econom, high-stress, track, moment, credit, corpor, secur, pressur, verg, employ, biggest, abil, demand, meet 
     Score: biggest, moment, environ, sourc, high-pressur, perform, stress, deadlin, often, whether, mistak, job, worri, pressur, meet, graduat, work, constant, well, enough, fast-pac, team, demand, expect, life, futur, bosss, abl, also, alway 
Topic 11 Top Words:
     Highest Prob: never, abl, find, job, ill, feel, ive, like, amount, start, look, depress, struggl, cycl, will, time, way, one, month, havent, unemploy, luck, neighborhood, improv, send, want, resum, odd, support, respect 
     FREX: never, amount, abl, luck, ill, find, look, havent, resum, unemploy, ive, month, cycl, odd, send, improv, start, neighborhood, job, gotten, interview, overcom, relief, forev, will, respect, suffer, wish, scrape, voic 
     Lift: gotten, interview, luck, resum, overcom, forev, odd, amount, havent, unemploy, send, improv, relief, month, suffer, opinion, neighborhood, endless, loan, scrape, elder, optimist, cycl, never, insid, voic, look, poor, abl, wonder 
     Score: never, gotten, find, abl, amount, luck, ill, ive, job, resum, start, havent, unemploy, cycl, look, month, will, send, neighborhood, interview, odd, improv, forev, overcom, struggl, suffer, told, scrape, respect, poor 
Topic 12 Top Words:
     Highest Prob: constant, racism, like, feel, black, discrimin, judg, alway, stress, woman, life, skin, exhaust, color, hard, prove, sourc, biggest, daili, basi, moment, treat, twice, seen, one, race, white, peopl, guard, equal 
     FREX: racism, color, skin, judg, discrimin, awar, basi, daili, prove, exhaust, counterpart, guard, race, equal, seen, black, safeti, unfair, twice, treat, watch, woman, target, face, depart, constant, opportun, experi, worth, stereotyp 
     Lift: avoid, imag, awar, base, confront, counterpart, crap, danger, fair, present, second-class, color, perceiv, convers, skin, exclud, found, oppress, stereotyp, unfair, judg, racism, target, basi, eggshel, daili, equal, guard, race, made 
     Score: racism, discrimin, color, avoid, black, skin, judg, twice, prove, basi, daili, biggest, seen, equal, awar, guard, moment, sourc, exhaust, race, counterpart, constant, unfair, safeti, target, watch, treat, woman, face, experi 
Topic 13 Top Words:
     Highest Prob: just, right, feel, realli, now, know, cant, struggl, dont, seem, anyth, like, everyth, time, tire, alon, depress, shake, lost, helpless, happi, hopeless, better, help, pointless, everyon, turn, true, need, fail 
     FREX: right, now, just, know, cant, shake, lost, helpless, seem, pointless, true, anyth, alon, hopeless, joy, realli, everyth, shouldnt, struggl, happi, better, motiv, help, wrong, dont, grate, funk, energi, tire, fail 
     Lift: ’ve, bring, pointless, joy, true, right, helpless, now, shake, lost, disappear, effort, dear, know, cant, hopeless, bleak, shouldnt, just, energi, seem, read, motiv, curl, blue, outgo, grate, better, seek, funk 
     Score: right, now, just, know, cant, realli, anyth, seem, struggl, shake, ’ve, dont, pointless, everyth, lost, helpless, tire, true, hopeless, turn, alon, joy, shouldnt, better, feel, help, motiv, fail, happi, funk 
Topic 14 Top Words:
     Highest Prob: thing, anoth, one, relationship, feel, last, lot, caus, like, constant, hard, make, alway, big, tri, fight, stress, save, pressur, someth, take, work-rel, sometim, health, togeth, stay, struggl, abl, deal, argu 
     FREX: anoth, big, relationship, last, strain, save, work-rel, complet, tension, argu, weve, one, togeth, thing, stay, debt, increas, disagr, hous, fight, patch, rough, partner, member, critic, caus, lack, stabil, readi, girlfriend 
     Lift: disagr, stabil, patch, rough, strain, critic, payment, big, tension, complet, increas, weve, lack, anoth, girlfriend, debt, hous, expens, death, cover, left, work-rel, argu, readi, decis, express, leav, hes, heart, griev 
     Score: stabil, anoth, last, relationship, thing, big, caus, work-rel, strain, complet, fight, argu, tension, disagr, one, weve, save, member, partner, patch, rough, critic, experi, hous, lack, expens, debt, stay, girlfriend, increas 
Topic 15 Top Words:
     Highest Prob: stress, main, realli, life, job, sourc, feel, work, late, just, constant, hard, day, start, offic, time, need, like, tire, peopl, surround, cut, posit, custom, today, overwhelm, focus, feet, keep, everi 
     FREX: main, offic, focus, cut, surround, servic, sourc, feet, stress, custom, posit, day, machin, mom, late, need, call, center, phone, today, life, work, job, start, nois, anymor, realli, cog, sleep, screw 
     Lift: nois, build, call, center, crazi, phone, stay--hom, focus, mom, servic, dread, machin, main, cog, offic, desk, feet, bore, posit, surround, burn, sleep, unmotiv, custom, challeng, concentr, drain, primari, store, consid 
     Score: main, sourc, stress, build, job, offic, realli, late, custom, work, start, life, day, servic, feet, surround, focus, phone, just, call, center, machin, constant, cut, tire, mom, dread, nois, posit, need 
Topic 16 Top Words:
     Highest Prob: feel, like, well, one, enough, pressur, succeed, expect, fall, make, thing, also, depress, school, good, famili, meet, work, lot, short, friend, behind, constant, compar, best, anoth, fact, even, miss, colleagu 
     FREX: succeed, well, fall, school, expect, short, pressur, compar, colleagu, behind, standard, one, best, though, boss, miss, friend, enough, meet, recognit, even, other, fact, criticis, like, product, feel, peer, academ, understand 
     Lift: criticis, product, peer, touch, girl, short, recognit, school, succeed, fall, standard, though, colleagu, academ, promot, best, well, attract, attent, notic, aspect, busi, behind, proper, father, expect, disappoint, boss, compar, high 
     Score: well, school, succeed, criticis, fall, expect, pressur, short, compar, enough, meet, colleagu, thing, though, behind, one, feel, boss, standard, friend, like, good, anoth, depress, best, disappoint, famili, also, make, troubl 
Topic 17 Top Words:
     Highest Prob: get, feel, like, stuck, dont, rut, work, know, job, ahead, hate, life, see, way, anywher, hard, depress, realli, seem, spin, today, wheel, dead-end, place, nowher, move, unhappi, chang, just, hold 
     FREX: stuck, rut, get, ahead, hate, dont, know, anywher, dead-end, see, spin, wheel, nowher, today, unhappi, advanc, prospect, low-pay, work, like, way, move, feel, chang, job, place, citi, predomin, away, seem 
     Lift: advanc, prospect, citi, low-pay, rut, dead-end, stuck, hate, ahead, nowher, get, unhappi, wheel, spin, capabl, anywher, heard, move, today, chang, result, predomin, away, dont, place, sick, see, area, hold, voic 
     Score: stuck, rut, get, citi, ahead, dont, know, hate, anywher, job, spin, dead-end, wheel, nowher, work, see, low-pay, like, unhappi, today, advanc, prospect, seem, way, move, feel, chang, hold, away, predomin 
Topic 18 Top Words:
     Highest Prob: realli, sure, what, late, feel, just, caus, think, want, hey, everyon, ive, might, anyon, someth, talk, white, share, depress, appreci, dont, advic, great, turn, earli, lot, seem, that, word, know 
     FREX: what, sure, advic, hey, word, late, appreci, great, thank, might, bed, encourag, earli, want, anyon, think, share, everyon, talk, guy, caus, realli, nice, listen, ive, morn, someth, white, snap, shake 
     Lift: word, advic, encourag, sort, wisdom, thank, stuff, nice, realiz, bed, twenti, morn, guy, great, listen, what, earli, appreci, snap, littl, hey, empti, chest, wors, sure, eat, share, mexico, might, honest 
     Score: what, hey, realli, sure, late, advic, sort, caus, word, might, share, just, ive, great, anyon, think, everyon, encourag, white, bed, thank, appreci, turn, earli, talk, want, wisdom, guy, nice, morn 
Topic 19 Top Words:
     Highest Prob: feel, make, thing, last, like, realli, first, dont, depress, second, lone, general, one, friend, anyon, sure, person, also, close, final, talk, current, use, peopl, that, lot, want, issu, just, career 
     FREX: second, first, last, general, final, close, use, lone, thing, make, climat, anyon, confid, career, talk, friend, singl, along, polit, weigh, self-imag, that, current, dont, progress, person, apart, disconnect, countri, despit 
     Lift: divid, unrest, disconnect, divis, insurmount, confid, despit, wont, self-imag, second, nearbi, final, first, close, climat, work-lif, scare, age, term, along, lone, general, last, offer, use, studi, progress, apart, singl, break- 
     Score: first, last, second, thing, lone, make, insurmount, close, general, final, climat, use, issu, anyon, dont, depress, person, realli, friend, polit, current, talk, singl, difficulti, countri, weigh, confid, disconnect, career, sure 
Topic 20 Top Words:
     Highest Prob: feel, like, tire, just, hard, black, matter, els, fight, depress, see, everyon, sometim, man, alway, get, constant, battl, peopl, uphil, strong, way, world, work, woman, tri, everi, twice, live, expect 
     FREX: matter, tire, els, battl, uphil, fight, black, half, strong, see, sometim, pretend, hard, poverti, escap, far, everyon, twice, okay, macho, man, everi, men, weak, stack, stoic, face, news, show, told 
     Lift: brave, deck, half, incred, stack, stoic, surviv, bear, inequ, macho, battl, poverti, uphil, pretend, act, matter, flail, strong, okay, els, weak, show, independ, told, far, fight, escap, weight, news, tire 
     Score: black, tire, matter, fight, incred, els, battl, uphil, half, strong, twice, man, see, just, everyon, told, pretend, weak, far, show, poverti, world, face, discrimin, disadvantag, hard, everi, okay, stack, macho 
Topic 21 Top Words:
     Highest Prob: realli, take, health, mental, toll, constant, stress, right, now, feel, life, start, much, tough, physic, sourc, can, demand, longer, keep, pressur, enjoy, restaur, alway, relax, hard, affect, cope, work, custom 
     FREX: take, toll, mental, health, longer, physic, restaur, demand, tough, much, start, yell, rude, realli, right, constant, now, affect, fast, free, impact, qualiti, enjoy, custom, rope, chanc, relax, stress, oblig, emot 
     Lift: chanc, rude, qualiti, oblig, yell, smile, rope, toll, longer, take, physic, restaur, industri, impact, mental, fast, health, free, pace, chaotic, tough, affect, food, busi, demand, breakdown, start, custom, ask, burn 
     Score: toll, take, mental, health, physic, chanc, realli, demand, stress, restaur, start, longer, sourc, now, right, constant, impact, custom, much, relax, tough, yell, rude, keep, cope, enjoy, affect, fast, free, can 
Topic 22 Top Words:
     Highest Prob: worri, futur, will, also, life, happen, sure, world, stress, state, live, sourc, feel, hold, biggest, whether, constant, moment, like, one, polic, current, day, someth, that, there, violenc, children, everi, abl 
     FREX: happen, hold, will, state, futur, polic, world, uncertainti, violenc, crime, worri, crossroad, pull, safe, whether, children, next, threat, direct, victim, there, paycheck, care, that, stop, current, everi, differ, sure, path 
     Lift: crossroad, economi, powerless, racist, uncertainti, crime, victim, pull, paycheck, grow, concern, happen, polic, safe, hold, mistreat, brutal, next, threat, violenc, scari, street, children, stop, stori, state, path, will, turmoil, direct 
     Score: economi, worri, will, polic, futur, happen, world, biggest, state, uncertainti, hold, crime, violenc, threat, moment, sourc, victim, children, crossroad, whether, safe, pull, direct, next, kill, stress, brutal, sure, paycheck, current 
Topic 23 Top Words:
     Highest Prob: lot, stress, can, deal, person, overwhelm, time, life, work, issu, anxieti, cope, difficult, one, caus, balanc, divorc, manag, differ, lead, someth, high, frustrat, probabl, put, quit, stressor, juggl, weigh, outsid 
     FREX: deal, lot, balanc, person, anxieti, difficult, issu, divorc, manag, cope, can, lead, overwhelm, worklif, major, stress, field, quit, process, high, overal, juggl, task, time, probabl, stressor, handl, outsid, differ, competit 
     Lift: task, major, field, process, worklif, overal, difficult, manag, balanc, divorc, anxieti, deal, competit, juggl, overwhelm, lead, cope, especi, issu, stressor, outsid, eas, handl, proper, million, high, latina, employ, primari, person 
     Score: major, stress, deal, overwhelm, can, issu, lot, difficult, person, cope, anxieti, balanc, lead, divorc, manag, field, caus, stressor, worklif, process, high, quit, juggl, probabl, competit, weigh, differ, work, outsid, time 
Topic 24 Top Words:
     Highest Prob: keep, cant, seem, break, everyth, like, tri, just, catch, alway, hard, head, water, feel, ever, your, behind, late, time, someth, tough, peopl, give, life, relax, work, best, someon, tread, around 
     FREX: break, keep, catch, seem, water, head, everyth, tri, your, cant, ever, tread, give, drown, pleas, behind, harder, schoolwork, done, best, energi, sibl, hard, week, someon, alway, verg, tough, drama, tabl 
     Lift: drama, drown, tabl, schoolwork, slowli, break, pleas, catch, water, head, your, keep, tread, harder, week, hospit, breakdown, done, outgo, tri, sibl, certain, store, student, give, sick, ever, colleg, harass, pile 
     Score: break, keep, catch, tabl, seem, cant, water, everyth, head, tri, your, just, alway, ever, drown, tread, harder, pleas, behind, drama, sibl, relax, done, hard, schoolwork, give, week, colleg, slowli, juggl 
Topic 25 Top Words:
     Highest Prob: feel, isol, overwhelm, may, depress, lone, anxious, like, thing, futur, mani, make, respons, sad, caus, past, pain, anyon, part, fatigu, recent, ill, whatev, understood, experienc, hopeless, physic, helpless, other, dont 
     FREX: isol, may, anxious, overwhelm, lone, past, sad, fatigu, pain, respons, understood, recent, part, whatev, feel, mani, experienc, futur, depress, anyon, helpless, goal, hopeless, thing, physic, other, loss, caus, make, financ 
     Lift: fatigu, isol, understood, past, may, anxious, pain, sad, condit, symptom, recent, whatev, overwhelm, experienc, lone, loss, financ, part, specif, goal, respons, worthless, physic, mani, disappoint, helpless, reason, contribut, hopeless, difficult 
     Score: fatigu, isol, may, overwhelm, lone, anxious, respons, sad, past, pain, whatev, feel, depress, mani, condit, physic, understood, thing, futur, experienc, recent, symptom, part, loss, caus, specif, financ, anyon, difficult, cope 
labelTopics(model_depression, n = 30)
Topic 1 Top Words:
     Highest Prob: feel, like, alway, perfect, time, make, never, good, mistak, enough, can, depress, cant, everyth, relax, allow, peopl, pressur, ever, human, measur, friend, let, love, respect, compar, alon, anyon, reach, famili 
     FREX: perfect, mistak, time, allow, alway, relax, human, never, respect, let, good, measur, make, like, ever, everyth, progress, feel, can, cant, enough, full, love, reach, peopl, daughter, less, worthi, mind, compar 
     Lift: allow, perfect, human, bodi, daughter, mind, respect, mistak, relax, notic, contact, capabl, attent, let, truli, cost, possibl, appear, sister, ever, time, progress, less, alway, exercis, never, full, measur, student, outsid 
     Score: allow, perfect, mistak, alway, never, relax, human, enough, cant, good, like, can, make, time, feel, everyth, measur, ever, depress, let, respect, pressur, compar, full, worthi, truli, friend, peopl, reach, progress 
Topic 2 Top Words:
     Highest Prob: struggl, feel, depress, job, find, also, famili, relationship, friend, health, enjoy, mental, like, provid, miss, partner, sever, home, behind, back, fail, doesnt, first, place, support, alon, adjust, move, rent, love 
     FREX: struggl, find, miss, enjoy, partner, back, friend, home, relationship, rent, homesick, adjust, famili, also, diagnos, fail, class, doesnt, sever, provid, foremost, job, depress, evict, health, mental, behind, move, profession, place 
     Lift: class, homesick, diagnos, rent, adjust, evict, foremost, symptom, part-tim, romant, profession, medic, back, struggl, partner, third, barrier, fun, find, asia, sever, miss, doesnt, home, enjoy, relationship, student, appear, move, fail 
     Score: struggl, class, find, job, relationship, depress, health, mental, famili, also, partner, friend, enjoy, homesick, provid, diagnos, miss, home, sever, fail, rent, first, back, adjust, symptom, evict, behind, foremost, move, profession 
Topic 3 Top Words:
     Highest Prob: job, long, hour, work, life, stress, sourc, also, biggest, worri, moment, lot, famili, paid, well, dont, get, respons, time, good, much, abl, constant, sometim, lose, futur, sure, say, support, fire 
     FREX: hour, long, paid, biggest, job, moment, sourc, work, respons, lose, also, sole, spend, breadwinn, life, laid, worri, microscop, fire, factori, immigr, much, afford, stress, well, injur, famili, lot, home, status 
     Lift: injur, visa, construct, worker, paid, sole, weekend, microscop, hour, breadwinn, english, long, respons, lose, spend, speak, factori, cost, afford, laid, sibl, immigr, communic, incom, carri, comfort, decis, goe, sight, healthi 
     Score: hour, long, biggest, sourc, paid, moment, job, stress, respons, worri, visa, work, also, well, life, lot, famili, immigr, breadwinn, lose, fire, status, sole, much, say, factori, microscop, injur, spend, abl 
Topic 4 Top Words:
     Highest Prob: feel, like, life, just, live, anyth, depress, motion, realli, potenti, purpos, dont, exist, want, fact, day, accomplish, without, peopl, wast, reach, look, forward, hobbi, real, passion, full, goal, differ, make 
     FREX: purpos, motion, accomplish, potenti, live, exist, without, anyth, forward, passion, life, real, wast, float, reach, hobbi, fact, meaning, drift, like, full, just, day, goal, feel, ambit, mean, fullest, connect, look 
     Lift: ambit, purpos, drift, meaning, passion, float, accomplish, forward, motion, without, fullest, real, space, exist, hobbi, potenti, wast, goal, full, reach, mean, live, kind, connect, noth, hes, necessarili, puppet, away, interest 
     Score: anyth, motion, ambit, purpos, potenti, live, life, just, exist, accomplish, wast, like, real, depress, feel, forward, meaning, float, reach, without, day, passion, full, fact, hobbi, drift, direct, look, anywher, goal 
Topic 5 Top Words:
     Highest Prob: realli, dont, think, man, asian, depress, feel, woman, know, like, white, way, belong, late, turn, help, alon, place, just, need, societi, hispan, world, cultur, there, anywher, fit, old, better, year 
     FREX: asian, man, belong, think, woman, turn, white, fit, old, cultur, societi, help, place, dont, sorri, dark, way, hello, born, alon, failur, know, anywher, whos, there, sound, need, realli, either, american 
     Lift: middl, privileg, whos, sound, american, asian, belong, old, fit, born, sexism, man, cultur, hello, sorri, group, either, dark, failur, bad, insid, think, societi, express, histori, woman, white, turn, place, mexico 
     Score: asian, man, privileg, think, belong, white, turn, know, dont, woman, realli, old, fit, cultur, societi, anywher, depress, dark, sorri, born, way, year, whos, help, hispan, middl, late, hello, failur, world 
Topic 6 Top Words:
     Highest Prob: worri, make, money, financi, end, meet, feel, pay, bill, situat, constant, famili, also, one, work, provid, abl, mani, futur, current, support, save, two, bare, insecur, polit, reli, communiti, woman, whether 
     FREX: end, money, pay, financi, worri, bill, meet, situat, make, reli, insecur, provid, bare, save, threaten, polit, two, mother, date, current, attack, support, pandem, famili, member, mani, constant, communiti, futur, one 
     Lift: pandem, date, reli, end, threaten, insecur, pay, bill, situat, money, financi, bare, full-tim, mother, burnt, save, latina, abil, cover, scrape, expens, yet, attack, member, ident, provid, recogn, either, two, meet 
     Score: money, financi, end, worri, pay, bill, meet, pandem, situat, make, provid, reli, bare, mani, insecur, mother, two, save, polit, current, constant, threaten, member, futur, abl, attack, famili, date, children, full-tim 
Topic 7 Top Words:
     Highest Prob: hispan, famili, alway, parent, realli, want, hard, live, everyon, get, come, hey, ive, share, now, work, two, expect, what, came, year, right, ago, kid, america, tradit, countri, just, grew, disappoint 
     FREX: parent, hispan, came, come, tradit, ago, grade, grew, kid, two, share, strict, want, countri, year, hey, famili, child, marri, everyon, deport, proud, immigr, america, household, felt, still, disappoint, older, live 
     Lift: adult, strict, undocu, came, didnt, household, shadow, stabl, tradit, child, brother, beat, older, grade, parent, grew, caught, ago, hispan, proud, marri, larg, educ, fulli, kid, sister, come, colleg, languag, deport 
     Score: hispan, undocu, parent, came, grew, tradit, hey, ago, share, strict, grade, year, two, alway, famili, deport, immigr, come, everyon, felt, want, countri, what, household, proud, america, status, child, ive, live 
Topic 8 Top Words:
     Highest Prob: enough, feel, like, good, depress, anyth, never, ill, success, smart, pretti, worth, happi, compar, measur, worthi, other, will, burden, everyon, women, deserv, love, talent, one, just, help, cut, thought, see 
     FREX: enough, smart, success, good, pretti, worthi, worth, anyth, talent, burden, measur, ill, women, deserv, skinni, like, compar, happi, feel, depress, other, handsom, never, thought, thin, beauti, screw, shouldnt, cut, nobodi 
     Lift: thin, smart, talent, skinni, handsom, burden, worthi, success, nobodi, worth, pretti, good, deserv, enough, women, thought, beauti, worthless, measur, screw, inferior, compet, ill, compar, other, awhil, shouldnt, attract, cut, anyth 
     Score: enough, good, smart, anyth, thin, success, like, worth, ill, depress, worthi, feel, measur, compar, burden, never, talent, skinni, pretti, women, deserv, screw, other, shouldnt, worthless, happi, will, thought, beauti, handsom 
Topic 9 Top Words:
     Highest Prob: thing, depress, feel, can, problem, mani, caus, combin, person, might, contribut, troubl, number, also, relationship, discourag, way, hopeless, often, mayb, life, reason, think, talk, factor, addit, therapist, due, may, help 
     FREX: problem, number, combin, mayb, contribut, factor, common, mani, includ, troubl, discourag, therapist, due, might, addit, thing, hopeless, can, isnt, pinpoint, reason, inadequaci, caus, whatev, trigger, often, person, difficulti, chronic, women 
     Lift: chronic, address, brain, common, doctor, lay, exact, extrem, number, includ, activ, factor, dishearten, mayb, problem, inadequaci, due, therapist, most, contribut, combin, isnt, pinpoint, trigger, mood, addit, usual, mani, condit, whatev 
     Score: problem, combin, address, thing, mani, contribut, common, depress, can, caus, mayb, number, therapist, troubl, might, includ, due, may, person, factor, whatev, chronic, addit, reason, pinpoint, relationship, often, lead, discourag, difficulti 
Topic 10 Top Words:
     Highest Prob: stress, sourc, work, moment, job, like, life, biggest, worri, feel, constant, pressur, also, environ, meet, often, enough, perform, whether, high-pressur, lot, alway, futur, abl, well, deadlin, expect, mistak, sometim, make 
     FREX: environ, moment, biggest, high-pressur, perform, sourc, often, pressur, whether, deadlin, meet, work, fast-pac, stress, worri, job, bosss, demand, mistak, retir, constant, team, fire, life, expect, demot, corpor, also, say, probabl 
     Lift: graduat, environ, fast-pac, high-pressur, scrutini, debilit, deadlin, perform, demot, team, often, bosss, retir, whether, second-guess, level, econom, high-stress, track, moment, credit, corpor, secur, pressur, verg, employ, biggest, abil, demand, meet 
     Score: biggest, moment, environ, sourc, high-pressur, perform, stress, deadlin, often, whether, mistak, job, worri, pressur, meet, graduat, work, constant, well, enough, fast-pac, team, demand, expect, life, futur, bosss, abl, also, alway 
Topic 11 Top Words:
     Highest Prob: never, abl, find, job, ill, feel, ive, like, amount, start, look, depress, struggl, cycl, will, time, way, one, month, havent, unemploy, luck, neighborhood, improv, send, want, resum, odd, support, respect 
     FREX: never, amount, abl, luck, ill, find, look, havent, resum, unemploy, ive, month, cycl, odd, send, improv, start, neighborhood, job, gotten, interview, overcom, relief, forev, will, respect, suffer, wish, scrape, voic 
     Lift: gotten, interview, luck, resum, overcom, forev, odd, amount, havent, unemploy, send, improv, relief, month, suffer, opinion, neighborhood, endless, loan, scrape, elder, optimist, cycl, never, insid, voic, look, poor, abl, wonder 
     Score: never, gotten, find, abl, amount, luck, ill, ive, job, resum, start, havent, unemploy, cycl, look, month, will, send, neighborhood, interview, odd, improv, forev, overcom, struggl, suffer, told, scrape, respect, poor 
Topic 12 Top Words:
     Highest Prob: constant, racism, like, feel, black, discrimin, judg, alway, stress, woman, life, skin, exhaust, color, hard, prove, sourc, biggest, daili, basi, moment, treat, twice, seen, one, race, white, peopl, guard, equal 
     FREX: racism, color, skin, judg, discrimin, awar, basi, daili, prove, exhaust, counterpart, guard, race, equal, seen, black, safeti, unfair, twice, treat, watch, woman, target, face, depart, constant, opportun, experi, worth, stereotyp 
     Lift: avoid, imag, awar, base, confront, counterpart, crap, danger, fair, present, second-class, color, perceiv, convers, skin, exclud, found, oppress, stereotyp, unfair, judg, racism, target, basi, eggshel, daili, equal, guard, race, made 
     Score: racism, discrimin, color, avoid, black, skin, judg, twice, prove, basi, daili, biggest, seen, equal, awar, guard, moment, sourc, exhaust, race, counterpart, constant, unfair, safeti, target, watch, treat, woman, face, experi 
Topic 13 Top Words:
     Highest Prob: just, right, feel, realli, now, know, cant, struggl, dont, seem, anyth, like, everyth, time, tire, alon, depress, shake, lost, helpless, happi, hopeless, better, help, pointless, everyon, turn, true, need, fail 
     FREX: right, now, just, know, cant, shake, lost, helpless, seem, pointless, true, anyth, alon, hopeless, joy, realli, everyth, shouldnt, struggl, happi, better, motiv, help, wrong, dont, grate, funk, energi, tire, fail 
     Lift: ’ve, bring, pointless, joy, true, right, helpless, now, shake, lost, disappear, effort, dear, know, cant, hopeless, bleak, shouldnt, just, energi, seem, read, motiv, curl, blue, outgo, grate, better, seek, funk 
     Score: right, now, just, know, cant, realli, anyth, seem, struggl, shake, ’ve, dont, pointless, everyth, lost, helpless, tire, true, hopeless, turn, alon, joy, shouldnt, better, feel, help, motiv, fail, happi, funk 
Topic 14 Top Words:
     Highest Prob: thing, anoth, one, relationship, feel, last, lot, caus, like, constant, hard, make, alway, big, tri, fight, stress, save, pressur, someth, take, work-rel, sometim, health, togeth, stay, struggl, abl, deal, argu 
     FREX: anoth, big, relationship, last, strain, save, work-rel, complet, tension, argu, weve, one, togeth, thing, stay, debt, increas, disagr, hous, fight, patch, rough, partner, member, critic, caus, lack, stabil, readi, girlfriend 
     Lift: disagr, stabil, patch, rough, strain, critic, payment, big, tension, complet, increas, weve, lack, anoth, girlfriend, debt, hous, expens, death, cover, left, work-rel, argu, readi, decis, express, leav, hes, heart, griev 
     Score: stabil, anoth, last, relationship, thing, big, caus, work-rel, strain, complet, fight, argu, tension, disagr, one, weve, save, member, partner, patch, rough, critic, experi, hous, lack, expens, debt, stay, girlfriend, increas 
Topic 15 Top Words:
     Highest Prob: stress, main, realli, life, job, sourc, feel, work, late, just, constant, hard, day, start, offic, time, need, like, tire, peopl, surround, cut, posit, custom, today, overwhelm, focus, feet, keep, everi 
     FREX: main, offic, focus, cut, surround, servic, sourc, feet, stress, custom, posit, day, machin, mom, late, need, call, center, phone, today, life, work, job, start, nois, anymor, realli, cog, sleep, screw 
     Lift: nois, build, call, center, crazi, phone, stay--hom, focus, mom, servic, dread, machin, main, cog, offic, desk, feet, bore, posit, surround, burn, sleep, unmotiv, custom, challeng, concentr, drain, primari, store, consid 
     Score: main, sourc, stress, build, job, offic, realli, late, custom, work, start, life, day, servic, feet, surround, focus, phone, just, call, center, machin, constant, cut, tire, mom, dread, nois, posit, need 
Topic 16 Top Words:
     Highest Prob: feel, like, well, one, enough, pressur, succeed, expect, fall, make, thing, also, depress, school, good, famili, meet, work, lot, short, friend, behind, constant, compar, best, anoth, fact, even, miss, colleagu 
     FREX: succeed, well, fall, school, expect, short, pressur, compar, colleagu, behind, standard, one, best, though, boss, miss, friend, enough, meet, recognit, even, other, fact, criticis, like, product, feel, peer, academ, understand 
     Lift: criticis, product, peer, touch, girl, short, recognit, school, succeed, fall, standard, though, colleagu, academ, promot, best, well, attract, attent, notic, aspect, busi, behind, proper, father, expect, disappoint, boss, compar, high 
     Score: well, school, succeed, criticis, fall, expect, pressur, short, compar, enough, meet, colleagu, thing, though, behind, one, feel, boss, standard, friend, like, good, anoth, depress, best, disappoint, famili, also, make, troubl 
Topic 17 Top Words:
     Highest Prob: get, feel, like, stuck, dont, rut, work, know, job, ahead, hate, life, see, way, anywher, hard, depress, realli, seem, spin, today, wheel, dead-end, place, nowher, move, unhappi, chang, just, hold 
     FREX: stuck, rut, get, ahead, hate, dont, know, anywher, dead-end, see, spin, wheel, nowher, today, unhappi, advanc, prospect, low-pay, work, like, way, move, feel, chang, job, place, citi, predomin, away, seem 
     Lift: advanc, prospect, citi, low-pay, rut, dead-end, stuck, hate, ahead, nowher, get, unhappi, wheel, spin, capabl, anywher, heard, move, today, chang, result, predomin, away, dont, place, sick, see, area, hold, voic 
     Score: stuck, rut, get, citi, ahead, dont, know, hate, anywher, job, spin, dead-end, wheel, nowher, work, see, low-pay, like, unhappi, today, advanc, prospect, seem, way, move, feel, chang, hold, away, predomin 
Topic 18 Top Words:
     Highest Prob: realli, sure, what, late, feel, just, caus, think, want, hey, everyon, ive, might, anyon, someth, talk, white, share, depress, appreci, dont, advic, great, turn, earli, lot, seem, that, word, know 
     FREX: what, sure, advic, hey, word, late, appreci, great, thank, might, bed, encourag, earli, want, anyon, think, share, everyon, talk, guy, caus, realli, nice, listen, ive, morn, someth, white, snap, shake 
     Lift: word, advic, encourag, sort, wisdom, thank, stuff, nice, realiz, bed, twenti, morn, guy, great, listen, what, earli, appreci, snap, littl, hey, empti, chest, wors, sure, eat, share, mexico, might, honest 
     Score: what, hey, realli, sure, late, advic, sort, caus, word, might, share, just, ive, great, anyon, think, everyon, encourag, white, bed, thank, appreci, turn, earli, talk, want, wisdom, guy, nice, morn 
Topic 19 Top Words:
     Highest Prob: feel, make, thing, last, like, realli, first, dont, depress, second, lone, general, one, friend, anyon, sure, person, also, close, final, talk, current, use, peopl, that, lot, want, issu, just, career 
     FREX: second, first, last, general, final, close, use, lone, thing, make, climat, anyon, confid, career, talk, friend, singl, along, polit, weigh, self-imag, that, current, dont, progress, person, apart, disconnect, countri, despit 
     Lift: divid, unrest, disconnect, divis, insurmount, confid, despit, wont, self-imag, second, nearbi, final, first, close, climat, work-lif, scare, age, term, along, lone, general, last, offer, use, studi, progress, apart, singl, break- 
     Score: first, last, second, thing, lone, make, insurmount, close, general, final, climat, use, issu, anyon, dont, depress, person, realli, friend, polit, current, talk, singl, difficulti, countri, weigh, confid, disconnect, career, sure 
Topic 20 Top Words:
     Highest Prob: feel, like, tire, just, hard, black, matter, els, fight, depress, see, everyon, sometim, man, alway, get, constant, battl, peopl, uphil, strong, way, world, work, woman, tri, everi, twice, live, expect 
     FREX: matter, tire, els, battl, uphil, fight, black, half, strong, see, sometim, pretend, hard, poverti, escap, far, everyon, twice, okay, macho, man, everi, men, weak, stack, stoic, face, news, show, told 
     Lift: brave, deck, half, incred, stack, stoic, surviv, bear, inequ, macho, battl, poverti, uphil, pretend, act, matter, flail, strong, okay, els, weak, show, independ, told, far, fight, escap, weight, news, tire 
     Score: black, tire, matter, fight, incred, els, battl, uphil, half, strong, twice, man, see, just, everyon, told, pretend, weak, far, show, poverti, world, face, discrimin, disadvantag, hard, everi, okay, stack, macho 
Topic 21 Top Words:
     Highest Prob: realli, take, health, mental, toll, constant, stress, right, now, feel, life, start, much, tough, physic, sourc, can, demand, longer, keep, pressur, enjoy, restaur, alway, relax, hard, affect, cope, work, custom 
     FREX: take, toll, mental, health, longer, physic, restaur, demand, tough, much, start, yell, rude, realli, right, constant, now, affect, fast, free, impact, qualiti, enjoy, custom, rope, chanc, relax, stress, oblig, emot 
     Lift: chanc, rude, qualiti, oblig, yell, smile, rope, toll, longer, take, physic, restaur, industri, impact, mental, fast, health, free, pace, chaotic, tough, affect, food, busi, demand, breakdown, start, custom, ask, burn 
     Score: toll, take, mental, health, physic, chanc, realli, demand, stress, restaur, start, longer, sourc, now, right, constant, impact, custom, much, relax, tough, yell, rude, keep, cope, enjoy, affect, fast, free, can 
Topic 22 Top Words:
     Highest Prob: worri, futur, will, also, life, happen, sure, world, stress, state, live, sourc, feel, hold, biggest, whether, constant, moment, like, one, polic, current, day, someth, that, there, violenc, children, everi, abl 
     FREX: happen, hold, will, state, futur, polic, world, uncertainti, violenc, crime, worri, crossroad, pull, safe, whether, children, next, threat, direct, victim, there, paycheck, care, that, stop, current, everi, differ, sure, path 
     Lift: crossroad, economi, powerless, racist, uncertainti, crime, victim, pull, paycheck, grow, concern, happen, polic, safe, hold, mistreat, brutal, next, threat, violenc, scari, street, children, stop, stori, state, path, will, turmoil, direct 
     Score: economi, worri, will, polic, futur, happen, world, biggest, state, uncertainti, hold, crime, violenc, threat, moment, sourc, victim, children, crossroad, whether, safe, pull, direct, next, kill, stress, brutal, sure, paycheck, current 
Topic 23 Top Words:
     Highest Prob: lot, stress, can, deal, person, overwhelm, time, life, work, issu, anxieti, cope, difficult, one, caus, balanc, divorc, manag, differ, lead, someth, high, frustrat, probabl, put, quit, stressor, juggl, weigh, outsid 
     FREX: deal, lot, balanc, person, anxieti, difficult, issu, divorc, manag, cope, can, lead, overwhelm, worklif, major, stress, field, quit, process, high, overal, juggl, task, time, probabl, stressor, handl, outsid, differ, competit 
     Lift: task, major, field, process, worklif, overal, difficult, manag, balanc, divorc, anxieti, deal, competit, juggl, overwhelm, lead, cope, especi, issu, stressor, outsid, eas, handl, proper, million, high, latina, employ, primari, person 
     Score: major, stress, deal, overwhelm, can, issu, lot, difficult, person, cope, anxieti, balanc, lead, divorc, manag, field, caus, stressor, worklif, process, high, quit, juggl, probabl, competit, weigh, differ, work, outsid, time 
Topic 24 Top Words:
     Highest Prob: keep, cant, seem, break, everyth, like, tri, just, catch, alway, hard, head, water, feel, ever, your, behind, late, time, someth, tough, peopl, give, life, relax, work, best, someon, tread, around 
     FREX: break, keep, catch, seem, water, head, everyth, tri, your, cant, ever, tread, give, drown, pleas, behind, harder, schoolwork, done, best, energi, sibl, hard, week, someon, alway, verg, tough, drama, tabl 
     Lift: drama, drown, tabl, schoolwork, slowli, break, pleas, catch, water, head, your, keep, tread, harder, week, hospit, breakdown, done, outgo, tri, sibl, certain, store, student, give, sick, ever, colleg, harass, pile 
     Score: break, keep, catch, tabl, seem, cant, water, everyth, head, tri, your, just, alway, ever, drown, tread, harder, pleas, behind, drama, sibl, relax, done, hard, schoolwork, give, week, colleg, slowli, juggl 
Topic 25 Top Words:
     Highest Prob: feel, isol, overwhelm, may, depress, lone, anxious, like, thing, futur, mani, make, respons, sad, caus, past, pain, anyon, part, fatigu, recent, ill, whatev, understood, experienc, hopeless, physic, helpless, other, dont 
     FREX: isol, may, anxious, overwhelm, lone, past, sad, fatigu, pain, respons, understood, recent, part, whatev, feel, mani, experienc, futur, depress, anyon, helpless, goal, hopeless, thing, physic, other, loss, caus, make, financ 
     Lift: fatigu, isol, understood, past, may, anxious, pain, sad, condit, symptom, recent, whatev, overwhelm, experienc, lone, loss, financ, part, specif, goal, respons, worthless, physic, mani, disappoint, helpless, reason, contribut, hopeless, difficult 
     Score: fatigu, isol, may, overwhelm, lone, anxious, respons, sad, past, pain, whatev, feel, depress, mani, condit, physic, understood, thing, futur, experienc, recent, symptom, part, loss, caus, specif, financ, anyon, difficult, cope 
labelTopics(model_depression, n = 30)
findThoughts(model_depression, texts = survey_meta$text,topics = 4)# plots cloud of most important words

 Topic 4: 
     

I am feeling depressed because I feel like I am not living up to my full potential. I feel like I am not doing anything with my life that is meaningful or important. I feel like I am just going through the motions day after day and that my life has no purpose.
    

I am depressed because I feel like I am not living up to my potential. I feel like I am not doing anything with my life that is meaningful or important. I feel like I am just going through the motions and not really living.
    

I am feeling depressed because I feel like I am not doing anything with my life. I feel like I am not accomplishing anything and that my life has no purpose.
labelTopics(model_depression, topics = 25, n=30)
Topic 25 Top Words:
     Highest Prob: feel, isol, overwhelm, may, depress, lone, anxious, like, thing, futur, mani, make, respons, sad, caus, past, pain, anyon, part, fatigu, recent, ill, whatev, understood, experienc, hopeless, physic, helpless, other, dont 
     FREX: isol, may, anxious, overwhelm, lone, past, sad, fatigu, pain, respons, understood, recent, part, whatev, feel, mani, experienc, futur, depress, anyon, helpless, goal, hopeless, thing, physic, other, loss, caus, make, financ 
     Lift: fatigu, isol, understood, past, may, anxious, pain, sad, condit, symptom, recent, whatev, overwhelm, experienc, lone, loss, financ, part, specif, goal, respons, worthless, physic, mani, disappoint, helpless, reason, contribut, hopeless, difficult 
     Score: fatigu, isol, may, overwhelm, lone, anxious, respons, sad, past, pain, whatev, feel, depress, mani, condit, physic, understood, thing, futur, experienc, recent, symptom, part, loss, caus, specif, financ, anyon, difficult, cope 
findThoughts(model_depression, texts = survey_meta$text,topics = 25)# plots cloud of most important words
Error in findThoughts(model_depression, texts = survey_meta$text, topics = 25) : 
  Number of provided texts and number of documents modeled do not match
prep_gender <- stm::estimateEffect(~ gender, 
                                   model_depression,
                                   meta = survey_meta[, c( "gender", "race")], uncertainty = "Global")

prep_eth <- stm::estimateEffect(~ race,
                                model_depression,
                                meta = survey_meta[, c( "gender", "race")], uncertainty = "Global")
prep_w_vs <- stm::estimateEffect(~ w_vs,
                                model_depression,
                                meta = survey_meta[, c( "gender", "race")], uncertainty = "Global")
Error in model.frame.default(termobj, data = metadata) : 
  object is not a matrix
# topics <- c("finances, relationships", "feeling stuck", "politics and news","talent and success?", "high pressure from work", "work life balance", "police brutality??", "stressors related to asian indiv", "finance and stress from job", "racism, discrimination", "feeling inadequate, comparing to others", "???-1", "loneliness and isolation","insecurities rooted in prejudice", "feeling aimless, lack of purpose", "family, stressors related to hispanic indiv", "providing for family", "words to describe mental state?", "mental health", "frustration, physical reactions", " mistakes, disappointment, probing your worth", "???-2", "stress related to manual labor", "Lack of control", "sexism, racism, money")
# topics <- c("making mistakes", "relationship", "working long hours", "helplessness", "race words", "finances - family", "family - tradition", "comparing to others", "???9", "pressure form job", "unemployment", "racism", "???13", "work relatonship", "ovewhelmed from work", "???16", "helplessness", "???18", "???19", "fighting discrimination", "physical stress", "concern for future", "stress", "overworked", "loneliness")
topics <- c("making mistakes", "relationship", "working long hours", "helplessness", "race words", "finances - family", "family - tradition", "comparing to others", "pressure form job", "unemployment", "racism",  "work relatonship", "ovewhelmed from work", "helplessness", "fighting discrimination", "physical stress", "concern for future", "stress", "overworked", "loneliness")
topic_names = c(1,2,3,4,5,6,7,8,10,11,12,14,15,17,20,21,22,23,24,25)
plot(model_depression, type = "summary", xlim = c(0, 0.3))

plot(prep_gender, covariate = "gender", model = model_depression, method = "difference", 
     cov.value1 = "Male", cov.value2 = "Female",
     xlab = "More Female ... More Male",
     labeltype = "custom",
     custom.labels = topics,
     xlim = c(-0.1, 0.1),
     topics = topic_names,
     main = "Effect of Gender")

NA
NA
plot(prep_eth, covariate = "race", model = model_depression, method = "difference", 
     cov.value1 = "White", cov.value2 = "PoC",
     xlab = "More PoC ... More White",
     labeltype = "custom",
     custom.labels = topics,
     xlim = c(-0.05, 0.05),
     ylim = c(0, 20),
     topics = topic_names,
     main = "Effect of Race")
Error in sims[1, ] : subscript out of bounds
plot(prep_eth, covariate = "race", model = model_depression, method = "difference", 
     cov.value1 = "white", cov.value2 = "black",
     xlab = "More African American ... More white",
     labeltype = "custom",
     custom.labels = topics,
     topics = topic_names,
     xlim = c(-0.20, 0.20),
     )

plot(prep_eth, covariate = "race", model = model_depression, method = "difference", 
     cov.value1 = "white", cov.value2 = "asian",
     xlab = "More Asian ... More white",
     labeltype = "custom",
     custom.labels = topics,
     topics = topic_names,
     xlim = c(-0.20, 0.20),
     )

plot(prep_eth, covariate = "race", model = model_depression, method = "difference", 
     cov.value1 = "white", cov.value2 = "hispanic",
     xlab = "More hispanic ... More white",
     labeltype = "custom",
     custom.labels = topics,
     topics = topic_names,
     xlim = c(-0.20, 0.20),
     )

NA
NA

This is an R Markdown Notebook. When you execute code within the notebook, the results appear beneath the code.

Try executing this chunk by clicking the Run button within the chunk or by placing your cursor inside it and pressing Cmd+Shift+Enter.

## Installs
install.packages("stm")
install.packages("quanteda")
install.packages("spacyr")
install.packages("wordcloud")

## Imports


library(jsonlite)
library(stm)
library(quanteda)
library(spacyr)
library(stringr)
library(wordcloud)
library(ggplot2)
gender_func <- function(gender_label){
  if(gender_label == "woman"){
    return("Female")
  } else if(gender_label == "man") {
    return("Male")
  } else {
    return("other")
  }
}

w_vs <- function(eth_label){
  if(eth_label == 'white'){
    return("White")
  } else {
    return("Not White")
  }
}
a_vs <- function(eth_label){
  if(eth_label == 'asian'){
    return("Asian")
  } else {
    return("Not Asian")
  }
}

b_vs <- function(eth_label){
  if(eth_label == 'black'){
    return("African American")
  } else {
    return("Not African American")
  }
}
h_vs <- function(eth_label){
  if(eth_label == 'hispanic'){
    return("Hispanic")
  } else {
    return("Not Hispanic")
  }
}
# survey_meta <- read.csv("/Users/shinkamori/Documents/Research/p2345.csv")
survey_meta <- read.csv("/Users/shinkamori/Documents/Research/p2_p4_p5_p6_p7.csv")
survey_meta <- read.csv("/Users/shinkamori/Documents/Research/p4_5_6_7.csv")
survey_meta$gender <- sapply(survey_meta$gender, gender_func)
# survey_meta$w_vs <- sapply(survey_meta$race, w_vs)
# survey_meta$a_vs <- sapply(survey_meta$race, a_vs)
# survey_meta$b_vs <- sapply(survey_meta$race, b_vs)
# survey_meta$h_vs <- sapply(survey_meta$race, h_vs)

# survey_meta <- survey_meta[, c("text", "gender", "race")]
summary(survey_meta)
head(survey_metagit merge origin)
sapply(lapply(survey_meta, unique), length)
processed <- textProcessor(survey_meta$text, metadata=survey_meta[, c( "gender", "race")])
out <- prepDocuments(processed$documents, processed$vocab, processed$meta)
sapply(lapply(processed$meta, unique), length)
temp_text <- survey_meta[-processed$docs.removed,]
length(temp_text)
model_depression_race <- stm::stm(
  documents = out_race$documents,
  K = 25,
  prevalence =~ gender + w_vs ,
  data=out_race$meta,
  vocab = out_race$vocab,
  max.em.its=75
)
model_depression_race_2 <- stm::stm(
  documents = out_race$documents,
  K = 25,
  prevalence =~ gender + race ,
  data=out_race$meta,
  vocab = out_race$vocab,
  max.em.its=75
)
model_depression <- stm::stm(
  documents = out$documents,
  K = 25,
  prevalence =~ gender + race,
  data=out$meta,
  vocab = out$vocab,
  max.em.its=75
)
labelTopics(model_depression_race_2, n = 30)
labelTopics(model_depression, n = 30)
labelTopics(model_depression, n = 30)
findThoughts(model_depression, texts = survey_meta$text,topics = 4)# plots cloud of most important words
labelTopics(model_depression, topics = 25, n=30)
findThoughts(model_depression, texts = survey_meta$text,topics = 25)# plots cloud of most important words
cloud(model_depression, 25)
# plotQuote(thoughts, width = 30, main = "Topic 2")
prep_gender <- stm::estimateEffect(~ gender, 
                                   model_depression,
                                   meta = survey_meta[, c( "gender", "race")], uncertainty = "Global")

prep_eth <- stm::estimateEffect(~ race,
                                model_depression,
                                meta = survey_meta[, c( "gender", "race")], uncertainty = "Global")
prep_w_vs <- stm::estimateEffect(~ w_vs,
                                model_depression,
                                meta = survey_meta[, c( "gender", "race")], uncertainty = "Global")
# topics <- c("finances, relationships", "feeling stuck", "politics and news","talent and success?", "high pressure from work", "work life balance", "police brutality??", "stressors related to asian indiv", "finance and stress from job", "racism, discrimination", "feeling inadequate, comparing to others", "???-1", "loneliness and isolation","insecurities rooted in prejudice", "feeling aimless, lack of purpose", "family, stressors related to hispanic indiv", "providing for family", "words to describe mental state?", "mental health", "frustration, physical reactions", " mistakes, disappointment, probing your worth", "???-2", "stress related to manual labor", "Lack of control", "sexism, racism, money")
# topics <- c("making mistakes", "relationship", "working long hours", "helplessness", "race words", "finances - family", "family - tradition", "comparing to others", "???9", "pressure form job", "unemployment", "racism", "???13", "work relatonship", "ovewhelmed from work", "???16", "helplessness", "???18", "???19", "fighting discrimination", "physical stress", "concern for future", "stress", "overworked", "loneliness")
topics <- c("making mistakes", "relationship", "working long hours", "helplessness", "race words", "finances - family", "family - tradition", "comparing to others", "pressure form job", "unemployment", "racism",  "work relatonship", "ovewhelmed from work", "helplessness", "fighting discrimination", "physical stress", "concern for future", "stress", "overworked", "loneliness")
topic_names = c(1,2,3,4,5,6,7,8,10,11,12,14,15,17,20,21,22,23,24,25)
plot(model_depression, type = "summary", xlim = c(0, 0.3))
plot(prep_gender, covariate = "gender", model = model_depression, method = "difference", 
     cov.value1 = "Male", cov.value2 = "Female",
     xlab = "More Female ... More Male",
     labeltype = "custom",
     custom.labels = topics,
     xlim = c(-0.1, 0.1),
     topics = topic_names,
     main = "Effect of Gender")
plot(prep_eth, covariate = "race", model = model_depression, method = "difference", 
     cov.value1 = "White", cov.value2 = "PoC",
     xlab = "More PoC ... More White",
     labeltype = "custom",
     custom.labels = topics,
     xlim = c(-0.05, 0.05),
     ylim = c(0, 20),
     topics = topic_names,
     main = "Effect of Race")
plot(prep_eth, covariate = "race", model = model_depression, method = "difference", 
     cov.value1 = "white", cov.value2 = "black",
     xlab = "More African American ... More white",
     labeltype = "custom",
     custom.labels = topics,
     topics = topic_names,
     xlim = c(-0.20, 0.20),
     )
plot(prep_eth, covariate = "race", model = model_depression, method = "difference", 
     cov.value1 = "white", cov.value2 = "asian",
     xlab = "More Asian ... More white",
     labeltype = "custom",
     custom.labels = topics,
     topics = topic_names,
     xlim = c(-0.20, 0.20),
     )
plot(prep_eth, covariate = "race", model = model_depression, method = "difference", 
     cov.value1 = "white", cov.value2 = "hispanic",
     xlab = "More hispanic ... More white",
     labeltype = "custom",
     custom.labels = topics,
     topics = topic_names,
     xlim = c(-0.20, 0.20),
     )
toLDAvis(model_depression, out$documents)
install.packages("LDAvis")
plot(prep_eth, covariate = "race", model = model_depression, method = "difference", 
     cov.value1 = "asian", cov.value2 = "black",
     xlab = "More African American ... More Asian",
     labeltype = "custom",
     custom.labels = topics,
     
     xlim = c(-0.20, 0.20),
     )

plot(prep_eth, covariate = "race", model = model_depression, method = "difference", 
     cov.value1 = "hispanic", cov.value2 = "black",
     xlab = "More African American ... More Hispanic",
     labeltype = "custom",
     custom.labels = topics,
     xlim = c(-0.20, 0.20),
     )

plot(prep_eth, covariate = "race", model = model_depression, method = "difference", 
     cov.value1 = "white", cov.value2 = "black",
     xlab = "More African American ... More White",
     labeltype = "custom",
     custom.labels = topics,
     xlim = c(-0.20, 0.20),
     )

NA
NA

This is an R Markdown Notebook. When you execute code within the notebook, the results appear beneath the code.

Try executing this chunk by clicking the Run button within the chunk or by placing your cursor inside it and pressing Cmd+Shift+Enter.

## Installs
install.packages("stm")
install.packages("quanteda")
install.packages("spacyr")
install.packages("wordcloud")

## Imports


library(jsonlite)
library(stm)
library(quanteda)
library(spacyr)
library(stringr)
library(wordcloud)
library(ggplot2)
gender_func <- function(gender_label){
  if(gender_label == "woman"){
    return("Female")
  } else if(gender_label == "man") {
    return("Male")
  } else {
    return("other")
  }
}

w_vs <- function(eth_label){
  if(eth_label == 'white'){
    return("White")
  } else {
    return("Not White")
  }
}
a_vs <- function(eth_label){
  if(eth_label == 'asian'){
    return("Asian")
  } else {
    return("Not Asian")
  }
}

b_vs <- function(eth_label){
  if(eth_label == 'black'){
    return("African American")
  } else {
    return("Not African American")
  }
}
h_vs <- function(eth_label){
  if(eth_label == 'hispanic'){
    return("Hispanic")
  } else {
    return("Not Hispanic")
  }
}
# survey_meta <- read.csv("/Users/shinkamori/Documents/Research/p2345.csv")
survey_meta <- read.csv("/Users/shinkamori/Documents/Research/p2_p4_p5_p6_p7.csv")
survey_meta <- read.csv("/Users/shinkamori/Documents/Research/p4_5_6_7.csv")
survey_meta$gender <- sapply(survey_meta$gender, gender_func)
# survey_meta$w_vs <- sapply(survey_meta$race, w_vs)
# survey_meta$a_vs <- sapply(survey_meta$race, a_vs)
# survey_meta$b_vs <- sapply(survey_meta$race, b_vs)
# survey_meta$h_vs <- sapply(survey_meta$race, h_vs)

# survey_meta <- survey_meta[, c("text", "gender", "race")]
summary(survey_meta)
head(survey_metagit merge origin)
sapply(lapply(survey_meta, unique), length)
processed <- textProcessor(survey_meta$text, metadata=survey_meta[, c( "gender", "race")])
out <- prepDocuments(processed$documents, processed$vocab, processed$meta)
sapply(lapply(processed$meta, unique), length)
temp_text <- survey_meta[-processed$docs.removed,]
length(temp_text)
model_depression_race <- stm::stm(
  documents = out_race$documents,
  K = 25,
  prevalence =~ gender + w_vs ,
  data=out_race$meta,
  vocab = out_race$vocab,
  max.em.its=75
)
model_depression_race_2 <- stm::stm(
  documents = out_race$documents,
  K = 25,
  prevalence =~ gender + race ,
  data=out_race$meta,
  vocab = out_race$vocab,
  max.em.its=75
)
model_depression <- stm::stm(
  documents = out$documents,
  K = 25,
  prevalence =~ gender + race,
  data=out$meta,
  vocab = out$vocab,
  max.em.its=75
)
labelTopics(model_depression_race_2, n = 30)
labelTopics(model_depression, n = 30)
labelTopics(model_depression, n = 30)
findThoughts(model_depression, texts = survey_meta$text,topics = 4)# plots cloud of most important words
labelTopics(model_depression, topics = 25, n=30)
findThoughts(model_depression, texts = survey_meta$text,topics = 25)# plots cloud of most important words
cloud(model_depression, 25)
# plotQuote(thoughts, width = 30, main = "Topic 2")
prep_gender <- stm::estimateEffect(~ gender, 
                                   model_depression,
                                   meta = survey_meta[, c( "gender", "race")], uncertainty = "Global")

prep_eth <- stm::estimateEffect(~ race,
                                model_depression,
                                meta = survey_meta[, c( "gender", "race")], uncertainty = "Global")
prep_w_vs <- stm::estimateEffect(~ w_vs,
                                model_depression,
                                meta = survey_meta[, c( "gender", "race")], uncertainty = "Global")
# topics <- c("finances, relationships", "feeling stuck", "politics and news","talent and success?", "high pressure from work", "work life balance", "police brutality??", "stressors related to asian indiv", "finance and stress from job", "racism, discrimination", "feeling inadequate, comparing to others", "???-1", "loneliness and isolation","insecurities rooted in prejudice", "feeling aimless, lack of purpose", "family, stressors related to hispanic indiv", "providing for family", "words to describe mental state?", "mental health", "frustration, physical reactions", " mistakes, disappointment, probing your worth", "???-2", "stress related to manual labor", "Lack of control", "sexism, racism, money")
# topics <- c("making mistakes", "relationship", "working long hours", "helplessness", "race words", "finances - family", "family - tradition", "comparing to others", "???9", "pressure form job", "unemployment", "racism", "???13", "work relatonship", "ovewhelmed from work", "???16", "helplessness", "???18", "???19", "fighting discrimination", "physical stress", "concern for future", "stress", "overworked", "loneliness")
topics <- c("making mistakes", "relationship", "working long hours", "helplessness", "race words", "finances - family", "family - tradition", "comparing to others", "pressure form job", "unemployment", "racism",  "work relatonship", "ovewhelmed from work", "helplessness", "fighting discrimination", "physical stress", "concern for future", "stress", "overworked", "loneliness")
topic_names = c(1,2,3,4,5,6,7,8,10,11,12,14,15,17,20,21,22,23,24,25)
plot(model_depression, type = "summary", xlim = c(0, 0.3))
plot(prep_gender, covariate = "gender", model = model_depression, method = "difference", 
     cov.value1 = "Male", cov.value2 = "Female",
     xlab = "More Female ... More Male",
     labeltype = "custom",
     custom.labels = topics,
     xlim = c(-0.1, 0.1),
     topics = topic_names,
     main = "Effect of Gender")
plot(prep_eth, covariate = "race", model = model_depression, method = "difference", 
     cov.value1 = "White", cov.value2 = "PoC",
     xlab = "More PoC ... More White",
     labeltype = "custom",
     custom.labels = topics,
     xlim = c(-0.05, 0.05),
     ylim = c(0, 20),
     topics = topic_names,
     main = "Effect of Race")
plot(prep_eth, covariate = "race", model = model_depression, method = "difference", 
     cov.value1 = "white", cov.value2 = "black",
     xlab = "More African American ... More white",
     labeltype = "custom",
     custom.labels = topics,
     topics = topic_names,
     xlim = c(-0.20, 0.20),
     )
plot(prep_eth, covariate = "race", model = model_depression, method = "difference", 
     cov.value1 = "white", cov.value2 = "asian",
     xlab = "More Asian ... More white",
     labeltype = "custom",
     custom.labels = topics,
     topics = topic_names,
     xlim = c(-0.20, 0.20),
     )
plot(prep_eth, covariate = "race", model = model_depression, method = "difference", 
     cov.value1 = "white", cov.value2 = "hispanic",
     xlab = "More hispanic ... More white",
     labeltype = "custom",
     custom.labels = topics,
     topics = topic_names,
     xlim = c(-0.20, 0.20),
     )

This is an R Markdown Notebook. When you execute code within the notebook, the results appear beneath the code.

Try executing this chunk by clicking the Run button within the chunk or by placing your cursor inside it and pressing Cmd+Shift+Enter.

## Installs
install.packages("stm")
install.packages("quanteda")
install.packages("spacyr")
install.packages("wordcloud")

## Imports


library(jsonlite)
library(stm)
library(quanteda)
library(spacyr)
library(stringr)
library(wordcloud)
library(ggplot2)
gender_func <- function(gender_label){
  if(gender_label == "woman"){
    return("Female")
  } else if(gender_label == "man") {
    return("Male")
  } else {
    return("other")
  }
}

w_vs <- function(eth_label){
  if(eth_label == 'white'){
    return("White")
  } else {
    return("Not White")
  }
}
a_vs <- function(eth_label){
  if(eth_label == 'asian'){
    return("Asian")
  } else {
    return("Not Asian")
  }
}

b_vs <- function(eth_label){
  if(eth_label == 'black'){
    return("African American")
  } else {
    return("Not African American")
  }
}
h_vs <- function(eth_label){
  if(eth_label == 'hispanic'){
    return("Hispanic")
  } else {
    return("Not Hispanic")
  }
}
# survey_meta <- read.csv("/Users/shinkamori/Documents/Research/p2345.csv")
survey_meta <- read.csv("/Users/shinkamori/Documents/Research/p2_p4_p5_p6_p7.csv")
survey_meta <- read.csv("/Users/shinkamori/Documents/Research/p4_5_6_7.csv")
survey_meta$gender <- sapply(survey_meta$gender, gender_func)
# survey_meta$w_vs <- sapply(survey_meta$race, w_vs)
# survey_meta$a_vs <- sapply(survey_meta$race, a_vs)
# survey_meta$b_vs <- sapply(survey_meta$race, b_vs)
# survey_meta$h_vs <- sapply(survey_meta$race, h_vs)

# survey_meta <- survey_meta[, c("text", "gender", "race")]
summary(survey_meta)
head(survey_metagit merge origin)
sapply(lapply(survey_meta, unique), length)
processed <- textProcessor(survey_meta$text, metadata=survey_meta[, c( "gender", "race")])
out <- prepDocuments(processed$documents, processed$vocab, processed$meta)
sapply(lapply(processed$meta, unique), length)
temp_text <- survey_meta[-processed$docs.removed,]
length(temp_text)
model_depression_race <- stm::stm(
  documents = out_race$documents,
  K = 25,
  prevalence =~ gender + w_vs ,
  data=out_race$meta,
  vocab = out_race$vocab,
  max.em.its=75
)
model_depression_race_2 <- stm::stm(
  documents = out_race$documents,
  K = 25,
  prevalence =~ gender + race ,
  data=out_race$meta,
  vocab = out_race$vocab,
  max.em.its=75
)
model_depression <- stm::stm(
  documents = out$documents,
  K = 25,
  prevalence =~ gender + race,
  data=out$meta,
  vocab = out$vocab,
  max.em.its=75
)
labelTopics(model_depression_race_2, n = 30)
labelTopics(model_depression, n = 30)
labelTopics(model_depression, n = 30)
findThoughts(model_depression, texts = survey_meta$text,topics = 4)# plots cloud of most important words
labelTopics(model_depression, topics = 25, n=30)
findThoughts(model_depression, texts = survey_meta$text,topics = 25)# plots cloud of most important words
cloud(model_depression, 25)
# plotQuote(thoughts, width = 30, main = "Topic 2")
prep_gender <- stm::estimateEffect(~ gender, 
                                   model_depression,
                                   meta = survey_meta[, c( "gender", "race")], uncertainty = "Global")

prep_eth <- stm::estimateEffect(~ race,
                                model_depression,
                                meta = survey_meta[, c( "gender", "race")], uncertainty = "Global")
prep_w_vs <- stm::estimateEffect(~ w_vs,
                                model_depression,
                                meta = survey_meta[, c( "gender", "race")], uncertainty = "Global")
# topics <- c("finances, relationships", "feeling stuck", "politics and news","talent and success?", "high pressure from work", "work life balance", "police brutality??", "stressors related to asian indiv", "finance and stress from job", "racism, discrimination", "feeling inadequate, comparing to others", "???-1", "loneliness and isolation","insecurities rooted in prejudice", "feeling aimless, lack of purpose", "family, stressors related to hispanic indiv", "providing for family", "words to describe mental state?", "mental health", "frustration, physical reactions", " mistakes, disappointment, probing your worth", "???-2", "stress related to manual labor", "Lack of control", "sexism, racism, money")
# topics <- c("making mistakes", "relationship", "working long hours", "helplessness", "race words", "finances - family", "family - tradition", "comparing to others", "???9", "pressure form job", "unemployment", "racism", "???13", "work relatonship", "ovewhelmed from work", "???16", "helplessness", "???18", "???19", "fighting discrimination", "physical stress", "concern for future", "stress", "overworked", "loneliness")
topics <- c("making mistakes", "relationship", "working long hours", "helplessness", "race words", "finances - family", "family - tradition", "comparing to others", "pressure form job", "unemployment", "racism",  "work relatonship", "ovewhelmed from work", "helplessness", "fighting discrimination", "physical stress", "concern for future", "stress", "overworked", "loneliness")
topic_names = c(1,2,3,4,5,6,7,8,10,11,12,14,15,17,20,21,22,23,24,25)
plot(model_depression, type = "summary", xlim = c(0, 0.3))
plot(prep_gender, covariate = "gender", model = model_depression, method = "difference", 
     cov.value1 = "Male", cov.value2 = "Female",
     xlab = "More Female ... More Male",
     labeltype = "custom",
     custom.labels = topics,
     xlim = c(-0.1, 0.1),
     topics = topic_names,
     main = "Effect of Gender")
plot(prep_eth, covariate = "race", model = model_depression, method = "difference", 
     cov.value1 = "White", cov.value2 = "PoC",
     xlab = "More PoC ... More White",
     labeltype = "custom",
     custom.labels = topics,
     xlim = c(-0.05, 0.05),
     ylim = c(0, 20),
     topics = topic_names,
     main = "Effect of Race")
plot(prep_eth, covariate = "race", model = model_depression, method = "difference", 
     cov.value1 = "black", cov.value2 = "white",
     xlab = "More White ... More African American",
     labeltype = "custom",
     custom.labels = topics,
     topics = topic_names,
     xlim = c(-0.20, 0.20),
     )

plot(prep_eth, covariate = "race", model = model_depression, method = "difference", 
     cov.value1 = "asian", cov.value2 = "white",
     xlab = "More White ... More Asian",
     labeltype = "custom",
     custom.labels = topics,
     topics = topic_names,
     xlim = c(-0.20, 0.20),
     )

plot(prep_eth, covariate = "race", model = model_depression, method = "difference", 
     cov.value1 = "hispanic", cov.value2 = "white",
     xlab = "More White ... More Hispanic",
     labeltype = "custom",
     custom.labels = topics,
     topics = topic_names,
     xlim = c(-0.20, 0.20),
     )

NA
NA
toLDAvis(model_depression, out$documents)
install.packages("LDAvis")
plot(prep_eth, covariate = "race", model = model_depression, method = "difference", 
     cov.value1 = "asian", cov.value2 = "black",
     xlab = "More African American ... More Asian",
     labeltype = "custom",
     custom.labels = topics,
     topics = topic_names,
     xlim = c(-0.20, 0.20),
     )

plot(prep_eth, covariate = "race", model = model_depression, method = "difference", 
     cov.value1 = "hispanic", cov.value2 = "black",
     xlab = "More African American ... More Hispanic",
     labeltype = "custom",
     custom.labels = topics,
     xlim = c(-0.20, 0.20),
     topics = topic_names
     )

plot(prep_eth, covariate = "race", model = model_depression, method = "difference", 
     cov.value1 = "white", cov.value2 = "black",
     xlab = "More African American ... More White",
     labeltype = "custom",
     custom.labels = topics,
     xlim = c(-0.20, 0.20),
     topics = topic_names
     )

NA
NA
topicQuality(
  model_depression,
  out$documents,
  xlab = "Semantic Coherence",
  ylab = "Exclusivity",
 labels = topics,
  M = 20
)

e Insert Chunk button on the toolbar or by pressing Cmd+Option+I.

When you save the notebook, an HTML file containing the code and output will be saved alongside it (click the Preview button or press Cmd+Shift+K to preview the HTML file).

The preview shows you a rendered HTML copy of the contents of the editor. Consequently, unlike Knit, Preview does not run any R code chunks. Instead, the output of the chunk when it was last run in the editor is displayed.

toLDAvis(model_depression, out$documents)
install.packages("LDAvis")
plot(prep_eth, covariate = "race", model = model_depression, method = "difference", 
     cov.value1 = "asian", cov.value2 = "hispanic",
     xlab = "More Hispanic ... More Asian",
     labeltype = "custom",
     custom.labels = topics,
     topics = topic_names,
     
     xlim = c(-0.20, 0.20),
     )

plot(prep_eth, covariate = "race", model = model_depression, method = "difference", 
     cov.value1 = "black", cov.value2 = "hispanic",
     xlab = "More Hispanic ... More African American",
     labeltype = "custom",
     custom.labels = topics,
     xlim = c(-0.20, 0.20),
     topics = topic_names,
     )

plot(prep_eth, covariate = "race", model = model_depression, method = "difference", 
     cov.value1 = "white", cov.value2 = "hispanic",
     xlab = "More Hispanic ... More White",
     labeltype = "custom",
     custom.labels = topics,
     xlim = c(-0.20, 0.20),
     topics = topic_names
     )

NA
NA
plot(prep_eth, covariate = "race", model = model_depression, method = "difference", 
     cov.value1 = "white", cov.value2 = "asian",
     xlab = "More Asian ... More White",
     labeltype = "custom",
     custom.labels = topics,
      topics = topic_names,
     xlim = c(-0.20, 0.20),
     )

plot(prep_eth, covariate = "race", model = model_depression, method = "difference", 
     cov.value1 = "black", cov.value2 = "asian",
     xlab = "More Asian ... More African American",
     labeltype = "custom",
     custom.labels = topics,
      topics = topic_names,
     xlim = c(-0.20, 0.20),
     )

plot(prep_eth, covariate = "race", model = model_depression, method = "difference", 
     cov.value1 = "hispanic", cov.value2 = "asian",
     xlab = "More Asian ... More Hispanic",
     labeltype = "custom",
     custom.labels = topics,
      topics = topic_names,
     xlim = c(-0.20, 0.20),
     )

(semanticCoherence(model_depression, out$documents, M = 30))
topicQuality(
  model_depression,
  out$documents,
  xlab = "Semantic Coherence",
  ylab = "Exclusivity",
 labels = topics,
  M = 20
)

e Insert Chunk button on the toolbar or by pressing Cmd+Option+I.

When you save the notebook, an HTML file containing the code and output will be saved alongside it (click the Preview button or press Cmd+Shift+K to preview the HTML file).

The preview shows you a rendered HTML copy of the contents of the editor. Consequently, unlike Knit, Preview does not run any R code chunks. Instead, the output of the chunk when it was last run in the editor is displayed.

topicQuality(
  model_depression,
  out$documents,
  xlab = "Semantic Coherence",
  ylab = "Exclusivity",
 labels = topics,
  M = 20
)

e Insert Chunk button on the toolbar or by pressing Cmd+Option+I.

When you save the notebook, an HTML file containing the code and output will be saved alongside it (click the Preview button or press Cmd+Shift+K to preview the HTML file).

The preview shows you a rendered HTML copy of the contents of the editor. Consequently, unlike Knit, Preview does not run any R code chunks. Instead, the output of the chunk when it was last run in the editor is displayed.

toLDAvis(model_depression, out$documents)
install.packages("LDAvis")
plot(prep_eth, covariate = "race", model = model_depression, method = "difference", 
     cov.value1 = "asian", cov.value2 = "black",
     xlab = "More African American ... More Asian",
     labeltype = "custom",
     custom.labels = topics,
     
     xlim = c(-0.10, 0.10),
     )
plot(prep_eth, covariate = "race", model = model_depression, method = "difference", 
     cov.value1 = "hispanic", cov.value2 = "black",
     xlab = "More African American ... More Hispanic",
     labeltype = "custom",
     custom.labels = topics,
     xlim = c(-0.10, 0.10),
     )
plot(prep_eth, covariate = "race", model = model_depression, method = "difference", 
     cov.value1 = "asian", cov.value2 = "hispanic",
     xlab = "More hispanic ... More Asian",
     labeltype = "custom",
     custom.labels = topics,
     xlim = c(-0.10, 0.10),
     )
(semanticCoherence(model_depression, out$documents, M = 30))
topicQuality(
  model_depression,
  out$documents,
  xlab = "Semantic Coherence",
  ylab = "Exclusivity",
 labels = topics,
  M = 20
)

e Insert Chunk button on the toolbar or by pressing Cmd+Option+I.

When you save the notebook, an HTML file containing the code and output will be saved alongside it (click the Preview button or press Cmd+Shift+K to preview the HTML file).

The preview shows you a rendered HTML copy of the contents of the editor. Consequently, unlike Knit, Preview does not run any R code chunks. Instead, the output of the chunk when it was last run in the editor is displayed.

---
title: "R Notebook"
output: html_notebook
---

This is an [R Markdown](http://rmarkdown.rstudio.com) Notebook. When you execute code within the notebook, the results appear beneath the code. 

Try executing this chunk by clicking the *Run* button within the chunk or by placing your cursor inside it and pressing *Cmd+Shift+Enter*. 

```{r}
## Installs
install.packages("stm")
install.packages("quanteda")
install.packages("spacyr")
install.packages("wordcloud")

## Imports


library(jsonlite)
library(stm)
library(quanteda)
library(spacyr)
library(stringr)
library(wordcloud)
library(ggplot2)

```

```{r}
gender_func <- function(gender_label){
  if(gender_label == "woman"){
    return("Female")
  } else if(gender_label == "man") {
    return("Male")
  } else {
    return("other")
  }
}

w_vs <- function(eth_label){
  if(eth_label == 'white'){
    return("White")
  } else {
    return("Not White")
  }
}
a_vs <- function(eth_label){
  if(eth_label == 'asian'){
    return("Asian")
  } else {
    return("Not Asian")
  }
}

b_vs <- function(eth_label){
  if(eth_label == 'black'){
    return("African American")
  } else {
    return("Not African American")
  }
}
h_vs <- function(eth_label){
  if(eth_label == 'hispanic'){
    return("Hispanic")
  } else {
    return("Not Hispanic")
  }
}
```


```{r}
# survey_meta <- read.csv("/Users/shinkamori/Documents/Research/p2345.csv")
survey_meta <- read.csv("/Users/shinkamori/Documents/Research/p2_p4_p5_p6_p7.csv")
survey_meta <- read.csv("/Users/shinkamori/Documents/Research/p4_5_6_7.csv")
survey_meta$gender <- sapply(survey_meta$gender, gender_func)
# survey_meta$w_vs <- sapply(survey_meta$race, w_vs)
# survey_meta$a_vs <- sapply(survey_meta$race, a_vs)
# survey_meta$b_vs <- sapply(survey_meta$race, b_vs)
# survey_meta$h_vs <- sapply(survey_meta$race, h_vs)

# survey_meta <- survey_meta[, c("text", "gender", "race")]
```
```{r}
summary(survey_meta)
```
```{r}
head(survey_metagit merge origin)
```

```{r}
sapply(lapply(survey_meta, unique), length)
```


```{r}
processed <- textProcessor(survey_meta$text, metadata=survey_meta[, c( "gender", "race")])
out <- prepDocuments(processed$documents, processed$vocab, processed$meta)
```
```{r}
sapply(lapply(processed$meta, unique), length)

```
```{r}
temp_text <- survey_meta[-processed$docs.removed,]

```
```{r}
length(temp_text)
```
```{r}
model_depression_race <- stm::stm(
  documents = out_race$documents,
  K = 25,
  prevalence =~ gender + w_vs ,
  data=out_race$meta,
  vocab = out_race$vocab,
  max.em.its=75
)


```
```{r}
model_depression_race_2 <- stm::stm(
  documents = out_race$documents,
  K = 25,
  prevalence =~ gender + race ,
  data=out_race$meta,
  vocab = out_race$vocab,
  max.em.its=75
)

```


```{r}
model_depression <- stm::stm(
  documents = out$documents,
  K = 25,
  prevalence =~ gender + race,
  data=out$meta,
  vocab = out$vocab,
  max.em.its=75
)
```
```{r}
labelTopics(model_depression_race_2, n = 30)

```
```{r}
labelTopics(model_depression, n = 30)

```

```{r}
labelTopics(model_depression, n = 30)

```
```{r}
findThoughts(model_depression, texts = survey_meta$text,topics = 4)# plots cloud of most important words

```

```{r}
labelTopics(model_depression, topics = 25, n=30)
findThoughts(model_depression, texts = survey_meta$text,topics = 25)# plots cloud of most important words
cloud(model_depression, 25)
# plotQuote(thoughts, width = 30, main = "Topic 2")
```
```{r}
prep_gender <- stm::estimateEffect(~ gender, 
                                   model_depression,
                                   meta = survey_meta[, c( "gender", "race")], uncertainty = "Global")

prep_eth <- stm::estimateEffect(~ race,
                                model_depression,
                                meta = survey_meta[, c( "gender", "race")], uncertainty = "Global")
prep_w_vs <- stm::estimateEffect(~ w_vs,
                                model_depression,
                                meta = survey_meta[, c( "gender", "race")], uncertainty = "Global")
```



```{r}
# topics <- c("finances, relationships", "feeling stuck", "politics and news","talent and success?", "high pressure from work", "work life balance", "police brutality??", "stressors related to asian indiv", "finance and stress from job", "racism, discrimination", "feeling inadequate, comparing to others", "???-1", "loneliness and isolation","insecurities rooted in prejudice", "feeling aimless, lack of purpose", "family, stressors related to hispanic indiv", "providing for family", "words to describe mental state?", "mental health", "frustration, physical reactions", " mistakes, disappointment, probing your worth", "???-2", "stress related to manual labor", "Lack of control", "sexism, racism, money")
# topics <- c("making mistakes", "relationship", "working long hours", "helplessness", "race words", "finances - family", "family - tradition", "comparing to others", "???9", "pressure form job", "unemployment", "racism", "???13", "work relatonship", "ovewhelmed from work", "???16", "helplessness", "???18", "???19", "fighting discrimination", "physical stress", "concern for future", "stress", "overworked", "loneliness")
topics <- c("making mistakes", "relationship", "working long hours", "helplessness", "race words", "finances - family", "family - tradition", "comparing to others", "pressure form job", "unemployment", "racism",  "work relatonship", "ovewhelmed from work", "helplessness", "fighting discrimination", "physical stress", "concern for future", "stress", "overworked", "loneliness")
topic_names = c(1,2,3,4,5,6,7,8,10,11,12,14,15,17,20,21,22,23,24,25)


```

```{r}
plot(model_depression, type = "summary", xlim = c(0, 0.3))
plot(prep_gender, covariate = "gender", model = model_depression, method = "difference", 
     cov.value1 = "Male", cov.value2 = "Female",
     xlab = "More Female ... More Male",
     labeltype = "custom",
     custom.labels = topics,
     xlim = c(-0.1, 0.1),
     topics = topic_names,
     main = "Effect of Gender")


```

```{r}
plot(prep_eth, covariate = "race", model = model_depression, method = "difference", 
     cov.value1 = "White", cov.value2 = "PoC",
     xlab = "More PoC ... More White",
     labeltype = "custom",
     custom.labels = topics,
     xlim = c(-0.05, 0.05),
     ylim = c(0, 20),
     topics = topic_names,
     main = "Effect of Race")


```

```{r}
plot(prep_eth, covariate = "race", model = model_depression, method = "difference", 
     cov.value1 = "white", cov.value2 = "black",
     xlab = "More African American ... More white",
     labeltype = "custom",
     custom.labels = topics,
     topics = topic_names,
     xlim = c(-0.20, 0.20),
     )
plot(prep_eth, covariate = "race", model = model_depression, method = "difference", 
     cov.value1 = "white", cov.value2 = "asian",
     xlab = "More Asian ... More white",
     labeltype = "custom",
     custom.labels = topics,
     topics = topic_names,
     xlim = c(-0.20, 0.20),
     )
plot(prep_eth, covariate = "race", model = model_depression, method = "difference", 
     cov.value1 = "white", cov.value2 = "hispanic",
     xlab = "More hispanic ... More white",
     labeltype = "custom",
     custom.labels = topics,
     topics = topic_names,
     xlim = c(-0.20, 0.20),
     )


```

---
title: "R Notebook"
output: html_notebook
---

This is an [R Markdown](http://rmarkdown.rstudio.com) Notebook. When you execute code within the notebook, the results appear beneath the code. 

Try executing this chunk by clicking the *Run* button within the chunk or by placing your cursor inside it and pressing *Cmd+Shift+Enter*. 

```{r}
## Installs
install.packages("stm")
install.packages("quanteda")
install.packages("spacyr")
install.packages("wordcloud")

## Imports


library(jsonlite)
library(stm)
library(quanteda)
library(spacyr)
library(stringr)
library(wordcloud)
library(ggplot2)

```

```{r}
gender_func <- function(gender_label){
  if(gender_label == "woman"){
    return("Female")
  } else if(gender_label == "man") {
    return("Male")
  } else {
    return("other")
  }
}

w_vs <- function(eth_label){
  if(eth_label == 'white'){
    return("White")
  } else {
    return("Not White")
  }
}
a_vs <- function(eth_label){
  if(eth_label == 'asian'){
    return("Asian")
  } else {
    return("Not Asian")
  }
}

b_vs <- function(eth_label){
  if(eth_label == 'black'){
    return("African American")
  } else {
    return("Not African American")
  }
}
h_vs <- function(eth_label){
  if(eth_label == 'hispanic'){
    return("Hispanic")
  } else {
    return("Not Hispanic")
  }
}
```


```{r}
# survey_meta <- read.csv("/Users/shinkamori/Documents/Research/p2345.csv")
survey_meta <- read.csv("/Users/shinkamori/Documents/Research/p2_p4_p5_p6_p7.csv")
survey_meta <- read.csv("/Users/shinkamori/Documents/Research/p4_5_6_7.csv")
survey_meta$gender <- sapply(survey_meta$gender, gender_func)
# survey_meta$w_vs <- sapply(survey_meta$race, w_vs)
# survey_meta$a_vs <- sapply(survey_meta$race, a_vs)
# survey_meta$b_vs <- sapply(survey_meta$race, b_vs)
# survey_meta$h_vs <- sapply(survey_meta$race, h_vs)

# survey_meta <- survey_meta[, c("text", "gender", "race")]
```
```{r}
summary(survey_meta)
```
```{r}
head(survey_metagit merge origin)
```

```{r}
sapply(lapply(survey_meta, unique), length)
```


```{r}
processed <- textProcessor(survey_meta$text, metadata=survey_meta[, c( "gender", "race")])
out <- prepDocuments(processed$documents, processed$vocab, processed$meta)
```
```{r}
sapply(lapply(processed$meta, unique), length)

```
```{r}
temp_text <- survey_meta[-processed$docs.removed,]

```
```{r}
length(temp_text)
```
```{r}
model_depression_race <- stm::stm(
  documents = out_race$documents,
  K = 25,
  prevalence =~ gender + w_vs ,
  data=out_race$meta,
  vocab = out_race$vocab,
  max.em.its=75
)


```
```{r}
model_depression_race_2 <- stm::stm(
  documents = out_race$documents,
  K = 25,
  prevalence =~ gender + race ,
  data=out_race$meta,
  vocab = out_race$vocab,
  max.em.its=75
)

```


```{r}
model_depression <- stm::stm(
  documents = out$documents,
  K = 25,
  prevalence =~ gender + race,
  data=out$meta,
  vocab = out$vocab,
  max.em.its=75
)
```
```{r}
labelTopics(model_depression_race_2, n = 30)

```
```{r}
labelTopics(model_depression, n = 30)

```

```{r}
labelTopics(model_depression, n = 30)

```
```{r}
findThoughts(model_depression, texts = survey_meta$text,topics = 4)# plots cloud of most important words

```

```{r}
labelTopics(model_depression, topics = 25, n=30)
findThoughts(model_depression, texts = survey_meta$text,topics = 25)# plots cloud of most important words
cloud(model_depression, 25)
# plotQuote(thoughts, width = 30, main = "Topic 2")
```
```{r}
prep_gender <- stm::estimateEffect(~ gender, 
                                   model_depression,
                                   meta = survey_meta[, c( "gender", "race")], uncertainty = "Global")

prep_eth <- stm::estimateEffect(~ race,
                                model_depression,
                                meta = survey_meta[, c( "gender", "race")], uncertainty = "Global")
prep_w_vs <- stm::estimateEffect(~ w_vs,
                                model_depression,
                                meta = survey_meta[, c( "gender", "race")], uncertainty = "Global")
```



```{r}
# topics <- c("finances, relationships", "feeling stuck", "politics and news","talent and success?", "high pressure from work", "work life balance", "police brutality??", "stressors related to asian indiv", "finance and stress from job", "racism, discrimination", "feeling inadequate, comparing to others", "???-1", "loneliness and isolation","insecurities rooted in prejudice", "feeling aimless, lack of purpose", "family, stressors related to hispanic indiv", "providing for family", "words to describe mental state?", "mental health", "frustration, physical reactions", " mistakes, disappointment, probing your worth", "???-2", "stress related to manual labor", "Lack of control", "sexism, racism, money")
# topics <- c("making mistakes", "relationship", "working long hours", "helplessness", "race words", "finances - family", "family - tradition", "comparing to others", "???9", "pressure form job", "unemployment", "racism", "???13", "work relatonship", "ovewhelmed from work", "???16", "helplessness", "???18", "???19", "fighting discrimination", "physical stress", "concern for future", "stress", "overworked", "loneliness")
topics <- c("making mistakes", "relationship", "working long hours", "helplessness", "race words", "finances - family", "family - tradition", "comparing to others", "pressure form job", "unemployment", "racism",  "work relatonship", "ovewhelmed from work", "helplessness", "fighting discrimination", "physical stress", "concern for future", "stress", "overworked", "loneliness")
topic_names = c(1,2,3,4,5,6,7,8,10,11,12,14,15,17,20,21,22,23,24,25)


```

```{r}
plot(model_depression, type = "summary", xlim = c(0, 0.3))
plot(prep_gender, covariate = "gender", model = model_depression, method = "difference", 
     cov.value1 = "Male", cov.value2 = "Female",
     xlab = "More Female ... More Male",
     labeltype = "custom",
     custom.labels = topics,
     xlim = c(-0.1, 0.1),
     topics = topic_names,
     main = "Effect of Gender")


```

```{r}
plot(prep_eth, covariate = "race", model = model_depression, method = "difference", 
     cov.value1 = "White", cov.value2 = "PoC",
     xlab = "More PoC ... More White",
     labeltype = "custom",
     custom.labels = topics,
     xlim = c(-0.05, 0.05),
     ylim = c(0, 20),
     topics = topic_names,
     main = "Effect of Race")


```

```{r}
plot(prep_eth, covariate = "race", model = model_depression, method = "difference", 
     cov.value1 = "white", cov.value2 = "black",
     xlab = "More African American ... More white",
     labeltype = "custom",
     custom.labels = topics,
     topics = topic_names,
     xlim = c(-0.20, 0.20),
     )
plot(prep_eth, covariate = "race", model = model_depression, method = "difference", 
     cov.value1 = "white", cov.value2 = "asian",
     xlab = "More Asian ... More white",
     labeltype = "custom",
     custom.labels = topics,
     topics = topic_names,
     xlim = c(-0.20, 0.20),
     )
plot(prep_eth, covariate = "race", model = model_depression, method = "difference", 
     cov.value1 = "white", cov.value2 = "hispanic",
     xlab = "More hispanic ... More white",
     labeltype = "custom",
     custom.labels = topics,
     topics = topic_names,
     xlim = c(-0.20, 0.20),
     )


```

```{r}
toLDAvis(model_depression, out$documents)
```
```{r}
install.packages("LDAvis")

```


```{r}
plot(prep_eth, covariate = "race", model = model_depression, method = "difference", 
     cov.value1 = "asian", cov.value2 = "black",
     xlab = "More African American ... More Asian",
     labeltype = "custom",
     custom.labels = topics,
     
     xlim = c(-0.20, 0.20),
     )
plot(prep_eth, covariate = "race", model = model_depression, method = "difference", 
     cov.value1 = "hispanic", cov.value2 = "black",
     xlab = "More African American ... More Hispanic",
     labeltype = "custom",
     custom.labels = topics,
     xlim = c(-0.20, 0.20),
     )
plot(prep_eth, covariate = "race", model = model_depression, method = "difference", 
     cov.value1 = "white", cov.value2 = "black",
     xlab = "More African American ... More White",
     labeltype = "custom",
     custom.labels = topics,
     xlim = c(-0.20, 0.20),
     )


```

---
title: "R Notebook"
output: html_notebook
---

This is an [R Markdown](http://rmarkdown.rstudio.com) Notebook. When you execute code within the notebook, the results appear beneath the code. 

Try executing this chunk by clicking the *Run* button within the chunk or by placing your cursor inside it and pressing *Cmd+Shift+Enter*. 

```{r}
## Installs
install.packages("stm")
install.packages("quanteda")
install.packages("spacyr")
install.packages("wordcloud")

## Imports


library(jsonlite)
library(stm)
library(quanteda)
library(spacyr)
library(stringr)
library(wordcloud)
library(ggplot2)

```

```{r}
gender_func <- function(gender_label){
  if(gender_label == "woman"){
    return("Female")
  } else if(gender_label == "man") {
    return("Male")
  } else {
    return("other")
  }
}

w_vs <- function(eth_label){
  if(eth_label == 'white'){
    return("White")
  } else {
    return("Not White")
  }
}
a_vs <- function(eth_label){
  if(eth_label == 'asian'){
    return("Asian")
  } else {
    return("Not Asian")
  }
}

b_vs <- function(eth_label){
  if(eth_label == 'black'){
    return("African American")
  } else {
    return("Not African American")
  }
}
h_vs <- function(eth_label){
  if(eth_label == 'hispanic'){
    return("Hispanic")
  } else {
    return("Not Hispanic")
  }
}
```


```{r}
# survey_meta <- read.csv("/Users/shinkamori/Documents/Research/p2345.csv")
survey_meta <- read.csv("/Users/shinkamori/Documents/Research/p2_p4_p5_p6_p7.csv")
survey_meta <- read.csv("/Users/shinkamori/Documents/Research/p4_5_6_7.csv")
survey_meta$gender <- sapply(survey_meta$gender, gender_func)
# survey_meta$w_vs <- sapply(survey_meta$race, w_vs)
# survey_meta$a_vs <- sapply(survey_meta$race, a_vs)
# survey_meta$b_vs <- sapply(survey_meta$race, b_vs)
# survey_meta$h_vs <- sapply(survey_meta$race, h_vs)

# survey_meta <- survey_meta[, c("text", "gender", "race")]
```
```{r}
summary(survey_meta)
```
```{r}
head(survey_metagit merge origin)
```

```{r}
sapply(lapply(survey_meta, unique), length)
```


```{r}
processed <- textProcessor(survey_meta$text, metadata=survey_meta[, c( "gender", "race")])
out <- prepDocuments(processed$documents, processed$vocab, processed$meta)
```
```{r}
sapply(lapply(processed$meta, unique), length)

```
```{r}
temp_text <- survey_meta[-processed$docs.removed,]

```
```{r}
length(temp_text)
```
```{r}
model_depression_race <- stm::stm(
  documents = out_race$documents,
  K = 25,
  prevalence =~ gender + w_vs ,
  data=out_race$meta,
  vocab = out_race$vocab,
  max.em.its=75
)


```
```{r}
model_depression_race_2 <- stm::stm(
  documents = out_race$documents,
  K = 25,
  prevalence =~ gender + race ,
  data=out_race$meta,
  vocab = out_race$vocab,
  max.em.its=75
)

```


```{r}
model_depression <- stm::stm(
  documents = out$documents,
  K = 25,
  prevalence =~ gender + race,
  data=out$meta,
  vocab = out$vocab,
  max.em.its=75
)
```
```{r}
labelTopics(model_depression_race_2, n = 30)

```
```{r}
labelTopics(model_depression, n = 30)

```

```{r}
labelTopics(model_depression, n = 30)

```
```{r}
findThoughts(model_depression, texts = survey_meta$text,topics = 4)# plots cloud of most important words

```

```{r}
labelTopics(model_depression, topics = 25, n=30)
findThoughts(model_depression, texts = survey_meta$text,topics = 25)# plots cloud of most important words
cloud(model_depression, 25)
# plotQuote(thoughts, width = 30, main = "Topic 2")
```
```{r}
prep_gender <- stm::estimateEffect(~ gender, 
                                   model_depression,
                                   meta = survey_meta[, c( "gender", "race")], uncertainty = "Global")

prep_eth <- stm::estimateEffect(~ race,
                                model_depression,
                                meta = survey_meta[, c( "gender", "race")], uncertainty = "Global")
prep_w_vs <- stm::estimateEffect(~ w_vs,
                                model_depression,
                                meta = survey_meta[, c( "gender", "race")], uncertainty = "Global")
```



```{r}
# topics <- c("finances, relationships", "feeling stuck", "politics and news","talent and success?", "high pressure from work", "work life balance", "police brutality??", "stressors related to asian indiv", "finance and stress from job", "racism, discrimination", "feeling inadequate, comparing to others", "???-1", "loneliness and isolation","insecurities rooted in prejudice", "feeling aimless, lack of purpose", "family, stressors related to hispanic indiv", "providing for family", "words to describe mental state?", "mental health", "frustration, physical reactions", " mistakes, disappointment, probing your worth", "???-2", "stress related to manual labor", "Lack of control", "sexism, racism, money")
# topics <- c("making mistakes", "relationship", "working long hours", "helplessness", "race words", "finances - family", "family - tradition", "comparing to others", "???9", "pressure form job", "unemployment", "racism", "???13", "work relatonship", "ovewhelmed from work", "???16", "helplessness", "???18", "???19", "fighting discrimination", "physical stress", "concern for future", "stress", "overworked", "loneliness")
topics <- c("making mistakes", "relationship", "working long hours", "helplessness", "race words", "finances - family", "family - tradition", "comparing to others", "pressure form job", "unemployment", "racism",  "work relatonship", "ovewhelmed from work", "helplessness", "fighting discrimination", "physical stress", "concern for future", "stress", "overworked", "loneliness")
topic_names = c(1,2,3,4,5,6,7,8,10,11,12,14,15,17,20,21,22,23,24,25)


```

```{r}
plot(model_depression, type = "summary", xlim = c(0, 0.3))
plot(prep_gender, covariate = "gender", model = model_depression, method = "difference", 
     cov.value1 = "Male", cov.value2 = "Female",
     xlab = "More Female ... More Male",
     labeltype = "custom",
     custom.labels = topics,
     xlim = c(-0.1, 0.1),
     topics = topic_names,
     main = "Effect of Gender")


```

```{r}
plot(prep_eth, covariate = "race", model = model_depression, method = "difference", 
     cov.value1 = "White", cov.value2 = "PoC",
     xlab = "More PoC ... More White",
     labeltype = "custom",
     custom.labels = topics,
     xlim = c(-0.05, 0.05),
     ylim = c(0, 20),
     topics = topic_names,
     main = "Effect of Race")


```

```{r}
plot(prep_eth, covariate = "race", model = model_depression, method = "difference", 
     cov.value1 = "white", cov.value2 = "black",
     xlab = "More African American ... More white",
     labeltype = "custom",
     custom.labels = topics,
     topics = topic_names,
     xlim = c(-0.20, 0.20),
     )
plot(prep_eth, covariate = "race", model = model_depression, method = "difference", 
     cov.value1 = "white", cov.value2 = "asian",
     xlab = "More Asian ... More white",
     labeltype = "custom",
     custom.labels = topics,
     topics = topic_names,
     xlim = c(-0.20, 0.20),
     )
plot(prep_eth, covariate = "race", model = model_depression, method = "difference", 
     cov.value1 = "white", cov.value2 = "hispanic",
     xlab = "More hispanic ... More white",
     labeltype = "custom",
     custom.labels = topics,
     topics = topic_names,
     xlim = c(-0.20, 0.20),
     )


```

---
title: "R Notebook"
output: html_notebook
---

This is an [R Markdown](http://rmarkdown.rstudio.com) Notebook. When you execute code within the notebook, the results appear beneath the code. 

Try executing this chunk by clicking the *Run* button within the chunk or by placing your cursor inside it and pressing *Cmd+Shift+Enter*. 

```{r}
## Installs
install.packages("stm")
install.packages("quanteda")
install.packages("spacyr")
install.packages("wordcloud")

## Imports


library(jsonlite)
library(stm)
library(quanteda)
library(spacyr)
library(stringr)
library(wordcloud)
library(ggplot2)

```

```{r}
gender_func <- function(gender_label){
  if(gender_label == "woman"){
    return("Female")
  } else if(gender_label == "man") {
    return("Male")
  } else {
    return("other")
  }
}

w_vs <- function(eth_label){
  if(eth_label == 'white'){
    return("White")
  } else {
    return("Not White")
  }
}
a_vs <- function(eth_label){
  if(eth_label == 'asian'){
    return("Asian")
  } else {
    return("Not Asian")
  }
}

b_vs <- function(eth_label){
  if(eth_label == 'black'){
    return("African American")
  } else {
    return("Not African American")
  }
}
h_vs <- function(eth_label){
  if(eth_label == 'hispanic'){
    return("Hispanic")
  } else {
    return("Not Hispanic")
  }
}
```


```{r}
# survey_meta <- read.csv("/Users/shinkamori/Documents/Research/p2345.csv")
survey_meta <- read.csv("/Users/shinkamori/Documents/Research/p2_p4_p5_p6_p7.csv")
survey_meta <- read.csv("/Users/shinkamori/Documents/Research/p4_5_6_7.csv")
survey_meta$gender <- sapply(survey_meta$gender, gender_func)
# survey_meta$w_vs <- sapply(survey_meta$race, w_vs)
# survey_meta$a_vs <- sapply(survey_meta$race, a_vs)
# survey_meta$b_vs <- sapply(survey_meta$race, b_vs)
# survey_meta$h_vs <- sapply(survey_meta$race, h_vs)

# survey_meta <- survey_meta[, c("text", "gender", "race")]
```
```{r}
summary(survey_meta)
```
```{r}
head(survey_metagit merge origin)
```

```{r}
sapply(lapply(survey_meta, unique), length)
```


```{r}
processed <- textProcessor(survey_meta$text, metadata=survey_meta[, c( "gender", "race")])
out <- prepDocuments(processed$documents, processed$vocab, processed$meta)
```
```{r}
sapply(lapply(processed$meta, unique), length)

```
```{r}
temp_text <- survey_meta[-processed$docs.removed,]

```
```{r}
length(temp_text)
```
```{r}
model_depression_race <- stm::stm(
  documents = out_race$documents,
  K = 25,
  prevalence =~ gender + w_vs ,
  data=out_race$meta,
  vocab = out_race$vocab,
  max.em.its=75
)


```
```{r}
model_depression_race_2 <- stm::stm(
  documents = out_race$documents,
  K = 25,
  prevalence =~ gender + race ,
  data=out_race$meta,
  vocab = out_race$vocab,
  max.em.its=75
)

```


```{r}
model_depression <- stm::stm(
  documents = out$documents,
  K = 25,
  prevalence =~ gender + race,
  data=out$meta,
  vocab = out$vocab,
  max.em.its=75
)
```
```{r}
labelTopics(model_depression_race_2, n = 30)

```
```{r}
labelTopics(model_depression, n = 30)

```

```{r}
labelTopics(model_depression, n = 30)

```
```{r}
findThoughts(model_depression, texts = survey_meta$text,topics = 4)# plots cloud of most important words

```

```{r}
labelTopics(model_depression, topics = 25, n=30)
findThoughts(model_depression, texts = survey_meta$text,topics = 25)# plots cloud of most important words
cloud(model_depression, 25)
# plotQuote(thoughts, width = 30, main = "Topic 2")
```
```{r}
prep_gender <- stm::estimateEffect(~ gender, 
                                   model_depression,
                                   meta = survey_meta[, c( "gender", "race")], uncertainty = "Global")

prep_eth <- stm::estimateEffect(~ race,
                                model_depression,
                                meta = survey_meta[, c( "gender", "race")], uncertainty = "Global")
prep_w_vs <- stm::estimateEffect(~ w_vs,
                                model_depression,
                                meta = survey_meta[, c( "gender", "race")], uncertainty = "Global")
```



```{r}
# topics <- c("finances, relationships", "feeling stuck", "politics and news","talent and success?", "high pressure from work", "work life balance", "police brutality??", "stressors related to asian indiv", "finance and stress from job", "racism, discrimination", "feeling inadequate, comparing to others", "???-1", "loneliness and isolation","insecurities rooted in prejudice", "feeling aimless, lack of purpose", "family, stressors related to hispanic indiv", "providing for family", "words to describe mental state?", "mental health", "frustration, physical reactions", " mistakes, disappointment, probing your worth", "???-2", "stress related to manual labor", "Lack of control", "sexism, racism, money")
# topics <- c("making mistakes", "relationship", "working long hours", "helplessness", "race words", "finances - family", "family - tradition", "comparing to others", "???9", "pressure form job", "unemployment", "racism", "???13", "work relatonship", "ovewhelmed from work", "???16", "helplessness", "???18", "???19", "fighting discrimination", "physical stress", "concern for future", "stress", "overworked", "loneliness")
topics <- c("making mistakes", "relationship", "working long hours", "helplessness", "race words", "finances - family", "family - tradition", "comparing to others", "pressure form job", "unemployment", "racism",  "work relatonship", "ovewhelmed from work", "helplessness", "fighting discrimination", "physical stress", "concern for future", "stress", "overworked", "loneliness")
topic_names = c(1,2,3,4,5,6,7,8,10,11,12,14,15,17,20,21,22,23,24,25)


```

```{r}
plot(model_depression, type = "summary", xlim = c(0, 0.3))
plot(prep_gender, covariate = "gender", model = model_depression, method = "difference", 
     cov.value1 = "Male", cov.value2 = "Female",
     xlab = "More Female ... More Male",
     labeltype = "custom",
     custom.labels = topics,
     xlim = c(-0.1, 0.1),
     topics = topic_names,
     main = "Effect of Gender")


```

```{r}
plot(prep_eth, covariate = "race", model = model_depression, method = "difference", 
     cov.value1 = "White", cov.value2 = "PoC",
     xlab = "More PoC ... More White",
     labeltype = "custom",
     custom.labels = topics,
     xlim = c(-0.05, 0.05),
     ylim = c(0, 20),
     topics = topic_names,
     main = "Effect of Race")


```

```{r}
plot(prep_eth, covariate = "race", model = model_depression, method = "difference", 
     cov.value1 = "black", cov.value2 = "white",
     xlab = "More White ... More African American",
     labeltype = "custom",
     custom.labels = topics,
     topics = topic_names,
     xlim = c(-0.20, 0.20),
     )
plot(prep_eth, covariate = "race", model = model_depression, method = "difference", 
     cov.value1 = "asian", cov.value2 = "white",
     xlab = "More White ... More Asian",
     labeltype = "custom",
     custom.labels = topics,
     topics = topic_names,
     xlim = c(-0.20, 0.20),
     )
plot(prep_eth, covariate = "race", model = model_depression, method = "difference", 
     cov.value1 = "hispanic", cov.value2 = "white",
     xlab = "More White ... More Hispanic",
     labeltype = "custom",
     custom.labels = topics,
     topics = topic_names,
     xlim = c(-0.20, 0.20),
     )


```

```{r}
toLDAvis(model_depression, out$documents)
```
```{r}
install.packages("LDAvis")

```


```{r}
plot(prep_eth, covariate = "race", model = model_depression, method = "difference", 
     cov.value1 = "asian", cov.value2 = "black",
     xlab = "More African American ... More Asian",
     labeltype = "custom",
     custom.labels = topics,
     topics = topic_names,
     xlim = c(-0.20, 0.20),
     )
plot(prep_eth, covariate = "race", model = model_depression, method = "difference", 
     cov.value1 = "hispanic", cov.value2 = "black",
     xlab = "More African American ... More Hispanic",
     labeltype = "custom",
     custom.labels = topics,
     xlim = c(-0.20, 0.20),
     topics = topic_names
     )
plot(prep_eth, covariate = "race", model = model_depression, method = "difference", 
     cov.value1 = "white", cov.value2 = "black",
     xlab = "More African American ... More White",
     labeltype = "custom",
     custom.labels = topics,
     xlim = c(-0.20, 0.20),
     topics = topic_names
     )


```



``````{r}
topicQuality(
  model_depression,
  out$documents,
  xlab = "Semantic Coherence",
  ylab = "Exclusivity",
 labels = topics,
  M = 20
)
```

e *Insert Chunk* button on the toolbar or by pressing *Cmd+Option+I*.

When you save the notebook, an HTML file containing the code and output will be saved alongside it (click the *Preview* button or press *Cmd+Shift+K* to preview the HTML file). 

The preview shows you a rendered HTML copy of the contents of the editor. Consequently, unlike *Knit*, *Preview* does not run any R code chunks. Instead, the output of the chunk when it was last run in the editor is displayed.





```{r}
toLDAvis(model_depression, out$documents)
```
```{r}
install.packages("LDAvis")

```


```{r}
plot(prep_eth, covariate = "race", model = model_depression, method = "difference", 
     cov.value1 = "asian", cov.value2 = "hispanic",
     xlab = "More Hispanic ... More Asian",
     labeltype = "custom",
     custom.labels = topics,
     topics = topic_names,
     
     xlim = c(-0.20, 0.20),
     )
plot(prep_eth, covariate = "race", model = model_depression, method = "difference", 
     cov.value1 = "black", cov.value2 = "hispanic",
     xlab = "More Hispanic ... More African American",
     labeltype = "custom",
     custom.labels = topics,
     xlim = c(-0.20, 0.20),
     topics = topic_names,
     )
plot(prep_eth, covariate = "race", model = model_depression, method = "difference", 
     cov.value1 = "white", cov.value2 = "hispanic",
     xlab = "More Hispanic ... More White",
     labeltype = "custom",
     custom.labels = topics,
     xlim = c(-0.20, 0.20),
     topics = topic_names
     )


```
```{r}
plot(prep_eth, covariate = "race", model = model_depression, method = "difference", 
     cov.value1 = "white", cov.value2 = "asian",
     xlab = "More Asian ... More White",
     labeltype = "custom",
     custom.labels = topics,
      topics = topic_names,
     xlim = c(-0.20, 0.20),
     )
plot(prep_eth, covariate = "race", model = model_depression, method = "difference", 
     cov.value1 = "black", cov.value2 = "asian",
     xlab = "More Asian ... More African American",
     labeltype = "custom",
     custom.labels = topics,
      topics = topic_names,
     xlim = c(-0.20, 0.20),
     )
plot(prep_eth, covariate = "race", model = model_depression, method = "difference", 
     cov.value1 = "hispanic", cov.value2 = "asian",
     xlab = "More Asian ... More Hispanic",
     labeltype = "custom",
     custom.labels = topics,
      topics = topic_names,
     xlim = c(-0.20, 0.20),
     )
```
```{r}
(semanticCoherence(model_depression, out$documents, M = 30))
```



``````{r}
topicQuality(
  model_depression,
  out$documents,
  xlab = "Semantic Coherence",
  ylab = "Exclusivity",
 labels = topics,
  M = 20
)
```

e *Insert Chunk* button on the toolbar or by pressing *Cmd+Option+I*.

When you save the notebook, an HTML file containing the code and output will be saved alongside it (click the *Preview* button or press *Cmd+Shift+K* to preview the HTML file). 

The preview shows you a rendered HTML copy of the contents of the editor. Consequently, unlike *Knit*, *Preview* does not run any R code chunks. Instead, the output of the chunk when it was last run in the editor is displayed.



``````{r}
topicQuality(
  model_depression,
  out$documents,
  xlab = "Semantic Coherence",
  ylab = "Exclusivity",
 labels = topics,
  M = 20
)
```

e *Insert Chunk* button on the toolbar or by pressing *Cmd+Option+I*.

When you save the notebook, an HTML file containing the code and output will be saved alongside it (click the *Preview* button or press *Cmd+Shift+K* to preview the HTML file). 

The preview shows you a rendered HTML copy of the contents of the editor. Consequently, unlike *Knit*, *Preview* does not run any R code chunks. Instead, the output of the chunk when it was last run in the editor is displayed.





```{r}
toLDAvis(model_depression, out$documents)
```
```{r}
install.packages("LDAvis")

```


```{r}
plot(prep_eth, covariate = "race", model = model_depression, method = "difference", 
     cov.value1 = "asian", cov.value2 = "black",
     xlab = "More African American ... More Asian",
     labeltype = "custom",
     custom.labels = topics,
     
     xlim = c(-0.10, 0.10),
     )
plot(prep_eth, covariate = "race", model = model_depression, method = "difference", 
     cov.value1 = "hispanic", cov.value2 = "black",
     xlab = "More African American ... More Hispanic",
     labeltype = "custom",
     custom.labels = topics,
     xlim = c(-0.10, 0.10),
     )
plot(prep_eth, covariate = "race", model = model_depression, method = "difference", 
     cov.value1 = "asian", cov.value2 = "hispanic",
     xlab = "More hispanic ... More Asian",
     labeltype = "custom",
     custom.labels = topics,
     xlim = c(-0.10, 0.10),
     )


```
```{r}
(semanticCoherence(model_depression, out$documents, M = 30))
```



``````{r}
topicQuality(
  model_depression,
  out$documents,
  xlab = "Semantic Coherence",
  ylab = "Exclusivity",
 labels = topics,
  M = 20
)
```

e *Insert Chunk* button on the toolbar or by pressing *Cmd+Option+I*.

When you save the notebook, an HTML file containing the code and output will be saved alongside it (click the *Preview* button or press *Cmd+Shift+K* to preview the HTML file). 

The preview shows you a rendered HTML copy of the contents of the editor. Consequently, unlike *Knit*, *Preview* does not run any R code chunks. Instead, the output of the chunk when it was last run in the editor is displayed.

